reproductive success and paternity concentration in …...reproductive success and paternity...
Post on 13-Jul-2020
6 Views
Preview:
TRANSCRIPT
Reproductive success and paternity
concentration in wild male orangutans
(genus Pongo)
MSc candidate: Ivan Lenzi
Supervisor: Dr. Michael Krützen
April 2014
Anthropological Institute and Museum
1
Abstract
Due to the unpredictable and extensive ranging behavior of male orangutans, only
little is known about the outcome of their reproductive strategies. Unlike other great
apes, orangutans do not live in stable social groups, but rather exhibit individual-based
fission-fusion sociality. There are two recognizable sexually mature male morphs,
which are physically and behaviorally distinct and pursue a different mating strategy.
The extent to which males can successfully monopolize the access to fertile females is
debated and, due to ecological factors, expected to vary considerably between
populations. Monopolization potential and reproductive skew have been so far
difficult to estimate from observations alone. We used genetic techniques and long-
term behavioral data to reconstruct pedigrees in two orangutan populations located
in Sumatra (Suaq Balimbing) and Borneo (Tuanan). Genotyping was performed for 27
microsatellite loci from non-invasively collected fecal samples and we assigned 10
paternities in Suaq Balimbing (N=55 unique individuals) and 8 in Tuanan (N=58 unique
individuals). In Suaq reproductive success is skewed and the dominant flanged male
obtains more paternities than subordinate flanged males or unflanged males. In
Tuanan paternities were assigned to 3 flanged males and one unflanged one. The
results of this study suggest that differences in the distribution of reproductive
success among males varies between two populations and that monopolization
potential for the dominant male in Sumatran populations might be higher than
expected. Our study therefore lends insight in to the effectiveness of mating strategies
of this elusive great ape.
2
Table of Contents
1. INTRODUCTION ..................................................................................................................... 4
1.1 Social systems ............................................................................................................... 4
1.1.1 Female social organization and social structure .......................................................... 4
1.1.2 Male social structure ............................................................................................ 5
1.2 Male mating strategies ................................................................................................. 6
1.3 Diversity of mating systems in Apes ............................................................................. 7
1.3.1 Genus Hylobates ................................................................................................... 7
1.3.2 Genus Gorilla ........................................................................................................ 7
1.3.3 Genus Pan ............................................................................................................. 8
1.4 Orangutans: The unique case of the solitary red ape................................................... 8
1.5 Orangutans: Two species and two distinct mating systems? ....................................... 9
1.5.1 The mating system of the Sumatran orangutan (Pongo abelii) .......................... 10
1.5.2 The mating system of the Bornean orangutan (Pongo pygmaeus) .................... 11
1.6 Previous studies of paternity in orangutans ............................................................... 12
1.7 Paternity assignments in wild animal populations ........................................................... 13
2 HYPOTHESES ....................................................................................................................... 16
3 METHODS ............................................................................................................................ 18
3.1 From fecal sample to ready-to-use DNA .................................................................... 18
3.1.1 Sampling ............................................................................................................. 18
3.1.2 Extraction ............................................................................................................ 18
3.1.3 DNA quantification ............................................................................................. 18
3.2 Microsatellite genotyping: From DNA to genetic data ............................................... 19
3.2.1 PCR Multiplex ...................................................................................................... 19
3.2.2 Allele scoring ....................................................................................................... 20
3.2.3 Consensus MACRO .............................................................................................. 20
3.2.4 Identity analysis .................................................................................................. 20
3.2.5 Usefulness of genetic markers ............................................................................ 21
3.2.6 Parentage analysis .............................................................................................. 21
3.2.7 Developmental status of the father and reproductive history of the mother ... 22
3.2.8 Estimate of paternity concentration................................................................... 22
4 RESULTS .............................................................................................................................. 23
4.1 Suaq ............................................................................................................................ 23
3
4.2 Tuanan ........................................................................................................................ 24
5 DISCUSSION ........................................................................................................................ 26
5.1 Reproductive success of male orangutans ................................................................. 26
5.1.1 Suaq .................................................................................................................... 26
5.1.2 Tuanan ................................................................................................................ 27
5.2 How do these results improve our understanding of orangutan’s mating system? .. 28
5.2.1 Interspecific differences ..................................................................................... 28
5.2.2 Reproductive success of the unflanged male ..................................................... 28
5.2.3 Evidence for infanticide in Sumatran orangutans? ............................................ 29
5.3 Limitations of this study and outlook ......................................................................... 31
6 AKNOWLEDGMENTS ........................................................................................................... 33
7 LITERATURE CITED .............................................................................................................. 33
8 APPENDIX ............................................................................................................................ 43
8.1 Appendix: Materials and methods................................................................................... 43
8.2 Appendix: Results ....................................................................................................... 44
8.2.1 DNA extraction .................................................................................................... 44
8.2.2 Genotyping ......................................................................................................... 47
8.2.3 Error rates ........................................................................................................... 47
8.2.4 Evaluation of the genetic markers ...................................................................... 47
8.2.5 Investigation of descriptive genetic parameters and analytical power.............. 48
8.2.6 Parentage analysis: Parameters used ................................................................. 52
8.2.7 Maternity: Suaq .................................................................................................. 54
8.2.8 Paternity concentration: Suaq ............................................................................ 54
8.2.9 Maternity: Tuanan .............................................................................................. 55
4
1. INTRODUCTION
1.1 Social systems In most primate species, females mainly compete over food resources while males compete
over the access to mates. The availability and distribution of these resources have therefore a
significant impact on the patterns of competition and on the social structure of most species
(van Schaik 1983). Two types of competition are predicted (Janson and van Schaik 1988);
scramble and contest competition. Scramble competition occurs when the limiting resource is
highly dispersed in small, low-quality patches or is relatively abundant but its usage cannot be
monopolized. In contest competition, the limiting resource is distributed in large, high-quality
patches that are easier to monopolize (Wrangham 1979). These types of competition can act
on both within-group and between-group levels and will likely affect various aspects of a
species social system (Sterk et al. 1997).
1.1.1 Female social organization and social structure Socio-ecological models predict that female social organization is mainly shaped by feeding
competition and predation pressure (Fig 1-1, Sterck et al. 1997, Janson and Goldsmith 1995).
High predation pressures usually favor larger group sizes (Isbel 1994), whereas within-group
contest competition for food favors the formation of coalitions, social bonds and steep
dominance hierarchies among females. Scramble competition, in contrast, favors the
formation of small groups, less steep (or absent) dominance hierarchies and reduced
territoriality (Sterck et al. 1997). These competitive regimes lead, in the majority of primate
species, to two categories of female relationships: Resident-nepotistic and dispersal-
egalitarian (Sterck et al 1997, Kappeler and van Schaik 2002). Sterck et al. tested the solidity of
the socio-ecological model for females and found that 27 primate species agree with the
predictions.
Figure 1-1: Schematic representation of the socio-ecological model for females (Sterck et al.
1997)
5
1.1.2 Male social structure Food distribution and predation risk determine the distribution of females. Male distribution
can also be affected by food distribution and predation pressure (van Schaik 1983). However,
intrasexual competition is a major predictor of male distribution (Wrangham 1979). Males
compete for the access to a valuable resource that limits their reproductive success. For
species in which males do not provide parental care for their offspring, the most valuable
resource is the access to fertile females (Wrangham 1979). The type and strength of
competition are thus predicted to shape the patterns of male association and the socio-sexual
strategies.
An indicator of the strength of male-male competition is the “operational sex ratio (OSR)”
(Emlen and Oring 1977, Kvarnemo and Ahnesjö 1996). This measure represents the
proportion of fertile females to fertile males. The availability of fertile females at any given
time is influenced by several factors, such as the number of females per group, the
reproductive rate of females, the degree of breeding seasonality, and the female’s behavior
(Clutton-Brock and Parker 1992). In most primate species, the OSR is skewed to females
(MItani et al. 1996). As OSR becomes more skewed, contest competition between males
increases (Mitani et al. 1996, Grant et al. 1995).
Monopolization potential is a major determinant of primate mating systems (Kappeler and
van Schaik 2002) and the OSR can be used to estimate the extent to which males can
monopolize the access to fertile females. Factors like the number of males in a group, the
level of gregariousness of females and the level of synchrony of their receptive periods are
tightly linked to the likelihood of males to maintain exclusive access to females (Ostner et al.
2008). When the likelihood of monopolizing the mating opportunities is very high, the
predicted social system has highly despotic ranks, intolerance between males and more or
less frequent violent takeovers. In contrast, when the access to mates cannot be
monopolized, less steep dominance hierarchy and, sometimes, male-male coalitions are
expected (e.g. Steenbeek et al. 2000, van Noorwijk and van Schaik 2001, Sprague 1992,
Takahata et al. 1999, Berard 1999).
The differences in mating systems can be reflected in patterns of reproductive success.
Paternity concentration (or reproductive skew) represents how reproductive success is
distributed among males. In despotic species, paternity is expected to be highly concentrated,
and in this case only a small proportion of males is reproductively very successful (e.g.
Nsubuga et al. 2008), while the majority of males has low or no reproductive success at all. In
species with more relaxed dominance among males, several males are successful, with a
6
reduced variability of reproductive success among males (e.g Strier et al. 2011, Kutsukake and
Nunn 2006 for a review using mating skew).
Fig 1-2: Schematic representation of the socio-ecological model for males (adapted from
Sterck et al. 1997).
1.2 Male mating strategies Group composition, strength of male-male competition and monopolization potential are
factors that influence the reproductive strategies that males employ (Kappeler and van Schaik
2002).
When possible, males try to limit the access to receptive females (e.g. Mass et al. 2009,
Setchel et al. 2005, Watts 1998). In despotic societies, subordinate males try to mate
opportunistically (e.g. Matsumoto-Oda 1999, Matsubara 2003), form friendships with females
(e.g. Massen et al. 2012, Palombit et al. 1997) or form coalitions against the top-ranking males
(e.g. Highham and Maestripieri 2010).
In more egalitarian societies, sperm competition is a widespread strategy (Møller 1988).
Males compete for fertilization by producing large amounts of sperm. This type of competition
usually leads to anatomical adaptations, such as large testicles and highly resilient sperm
(Anderson and Dixon 2002).
7
Another factor linked to male-male competition and the evolution of alternative mating
strategies in primates, albeit rare, is male bimaturism. Male Mandrills (Mandrillus sphinx)
show reversible bimaturism that reflects dominance (Dixon et al. 1993). Dominant “fatted”
males sire most of the offspring in the group and subordinate “non-fatted” males mate
opportunistically (Wickings et al. 2009). Adult, sexually mature males of long-tailed macaques
(Macaca fascicularis) are present in two sexually mature forms. The transition to the one with
fully developed secondary sexual characteristics is irreversible but also flexible (Not sure what
you mean by ‘flexible’), with males completing their development following social as well as
internal clues (Van Noordwijk and van Schaik 2001).
1.3 Diversity of mating systems in Apes
1.3.1 Genus Hylobates Within the genus Hylobates, white-handed gibbons (Hylobates lar) are a highly territorial, pair-
bonded, family-living species (Brockelman et al. 1998). Siamangs (Symphalangus spp.) also live
in pairs but have also been observed living in groups with more than two adults with one
female and multiple males (Lappan 2007).
Several hypotheses for the evolution of social monogamy in Hylobates species have been
tested and it has been proposed that the main role of the male in the pair is the protection of
the offspring against other males (van Schaik and Dunbar 1990, Borries et al. 2011).
Recently, however, the strictness of monogamy has been questioned, as extra-group
copulations and partner changes appear to occur more often than previously thought
(Reichard and Sommer 1997).
1.3.2 Genus Gorilla Gorillas (Gorilla spp.) form single-male, multi-female groups consisting usually of 8-10
individuals, even though larger groups (up to 40 individuals) have been reported (Doran &
Mcneilage 1998). The low levels of sociality and a weak dominance hierarchy among females
observed within groups can be explained by the reduced intensity of feeding contest
competition among them (Watts 2001).
Female aggregate in order to benefit from the protection of a single dominant male (Watts
1989, 1991, Lucas et al. 2010). For this reason, the competition for the role of the local
dominant male (the so-called alpha silverback) is extremely high. Selection for large body size
and competitive ability in males is reflected in the extreme body size sexual dimorphism as
well as in canine size sexual dimorphism (Plavcan 2001). These features suggest that male
gorillas evolved those traits in order to outcompete other males. Paternity for the groups is
8
generally very highly concentrated on the local dominant male, even in presence of other
silverbacks (Bradley 2004, 2005, Nsubuga 2008).
1.3.3 Genus Pan Chimpanzees (Pan troglodytes) adopt a flexible fission-fusion social system (Lehmann and
Boesch 2004). Chimpanzee communities can have between 20 and 150 individuals and are
further divided in smaller subgroups as typical for fission-fusion societies (Goodall 1986).
Strong male cohesion and philopatry are observed and within each community. However,
both sexes have a dominance hierarchy, with the highest ranking individuals having priority of
access to the monopolizable resources (Goodall 1986). High-ranked females can access to
superior food resources and this positively affects their reproductive success (Pusey et al.
1997).
Several males live in each community and therefore the probability of monopolizing the
access to females is reduced (Wroblewski et al. 2009). Competition among males is therefore
based on a “priority of access” model (Altmann 1962). High-ranking males mate more often
with females and subordinate males can attempt to sire offspring through sperm competition
by mating opportunistically or by consorting with females (Tutin 1979). Paternity analysis
suggests that male success mostly reflects dominance hierarchy (Boesch et al. 2006).
However, deviation from the “priority of access” model has been reported. Subordinate,
young males are more successful than expected, and the extent of this deviation is likely to be
linked to the number of competitors, number of simultaneously receptive females, female
choice and the length of the periovulatory phase (Wroblewski et al. 2009).
Bonobos also live in multi-male, multi-female groups but the dominance hierarchy appears to
be more relaxed; females also form strong coalitions and have higher social status than males
(De Waal 1995). Promiscuity is the norm and several males mate with the local females.
Estrous females tend to be monopolized by high-rank males, even though the only paternity
analysis for wild bonobos revealed that most males are reproductively successful. Paternity
success for the dominant male was estimated at 0-40% for the highest ranking males (Gerloff
1999). However, due to uncertainty about the dominance status of the males at the time of
conception for several offspring, the interpretation of these results is problematic.
1.4 Orangutans: The unique case of the solitary red ape Unlike gibbons and the African great apes, orangutans (Pongo pygmaeus and Pongo abelii)
live non-gregariously in individual-based fission-fusion societies (van Schaik 1999, Galdikas
1985). Females show a high degree of philopatry (Arora et al. 2012, Morrogh-Bernard et al.
9
2011) and despite their solitary nature, seem to receive social benefits in doing so (van
Noordwijk et al. 2012). Similar to most mammal species, males tend to leave the natal area
once reached sexual maturity, as a response to intrasexual competition and inbreeding risk
(Arora et al. 2012, Morrogh-Bernard 2011 et al., Lawson et al. 2007).
In both recognized orangutan species there is flexible and irreversible male bimaturism. Adult,
sexually mature males are present in two morphs and it can take 20 or more years after
reaching adulthood before the transition. The changes that occur during the irreversible
transition from the unflanged morph to the flanged morph are quite drastic, with an increase
in body size and the development of secondary sexual characteristics such as more developed
throat sacs, wider cheek pads and longer hair on their backs (MacKinnon 1974). Throat sacks
and check pads are used by flanged males to emit and precisely direct a long-distance
vocalization, the long call (MacKinnon 1974). Long calls are used in social contest to attract
mates and to regulate the spatial organization of males (Delgado et al. 2009).
1.5 Orangutans: Two species and two distinct mating systems? In both orangutan species, female orangutans have an exceptionally slow life history (Wich et
al. 2004, Galdikas and Ashbury 2012) and live at relatively low densities (Husson et al. 2009).
These conditions generate highly skewed operational sex ratio (Mitani et al. 1996), with
several males competing over few fertile females. Sexual selection theory suggests that in
case of contest competition, the competing sex will evolve traits that enhance competing
abilities, such as large size, weaponry, threat signals, female harassment as well as alternative
reproductive strategies (Anderson and Iwasa 1996). Orangutans have extreme body size
sexual dimorphism as well as strong canine sexual dimorphism (Plavcan and van Schaik 1992,
1997). Both traits are often associated with contest competition and are observed in other
primate species that experience strong male-male contest competition (Plavcan and van
Schaik 1992).
Field observations strongly suggest that socio-ecological differences between the two species
are present and influence the mating system and patterns of competition among males
(Delgado and van Schaik 2000, Dunkel et al. 2013). In particular, food abundance is a well-
documented source of differentiation between the two species (Marshall et al. 2009). On
Sumatra food availability is higher compared to Borneo and this allows orangutans to live at
higher densities (Husson et al. 2009, Singleton et al. 2009). Females’ home range overlap is
also more expensive (Singleton et al. 2009). Additionally, males and females can associate for
longer periods of time and in larger parties (Delgado and van Schaik 2000, Mitra Setia et al.
10
2009). As predicted by socio-ecological models (fig 1-2), food availability and the level of
gregariousness of females are major predictors of male-male competition and monopolization
potential (Kappeler and van Schaik 2002). It is suggested that the differences in
gregariousness affect several components of the two orangutan species’ mating systems
(Delgado and van Schaik 2000). In order to take on a comparative study of male reproductive
success, it is fundamental to understand the differences in the mating system of these
species.
1.5.1 The mating system of the Sumatran orangutan (Pongo abelii) Sumatran orangutans live in rich environments that allow them to have higher levels of
gregariousness compared to Bornean orangutans (Husson et al. 2009). Party size is also
considerably larger (Mitra Setia et al. 2009). As females tend to be quite gregarious, a single
flanged male can associate with several females for prolonged periods of time (Singleton and
van Schaik 2002). It is therefore likely that a male can successfully monopolize the access to
several females (Pradhan et al. 2012).Field observations suggest indeed that with one single
male can dominate over a given territory (Utami et al. 2002, Singleton and van Schaik 2002).
The dominance hierarchy among males can be very steep and the top ranking male can
dominate for over a decade (Utami et al. 2002).
The dominant male employs the “sit-and-wait” reproductive strategy (Utami et al. 2002). He
emits long calls and attract receptive females. This strategy relies on the female preference
for flanged males. Playback experiments showed that females indeed approach spontaneous
long calls of the dominant flanged male (Delgado 2003, Mitra Setia and van Schaik 2007,
Utami and Mitra setia 1995). The fact that even non-receptive females with dependent
offspring approach the dominant male’s long call suggest that females rely on him for
protection (Mitra Setia and van Schaik 2007). In addition, forced copulations involving the
dominant male are never observed (Singleton and van Schaik 2002).
Unflanged males employ a different reproductive strategy, the “go-and-search” strategy
(Utami et al. 2002). Unflanged males cannot produce long calls while roaming the forest in
search for receptive females. Roaming unflanged males have been observed copulating,
suggesting that the go-and-search strategy might indeed facilitate the access to females.
Flanged males end the association between unflanged males and females by chasing the male
away (Schürmann 1981, Utami and Mitra-Setia 1995). However, unflanged males, being more
agile than flanged males, have never been observed getting injured by flanged males (Utami
Atmoko et al. 2009a).
11
It has been suggested that because of the high food availability in Sumatra, the sit-and-wait
strategy might only be effective for the dominant male (Pradhan and van Schaik 2012). He can
associate with several females for prolonged periods of time and successfully exclude other
flanged males. In addition, the local dominant flanged male is more likely to attract females
with his long calls (Mitra Setia et al. 2009).For subordinate flanged males, the sit-and-wait
strategy might also be unsuccessful because long calls advertise the presence to the local
dominant male, who is likely to approach other male’s calls and chase them away (Utami
Atmoko et al 2009a).The dominant male is in fact highly intolerant towards other flanged
males.
As a result of stable dominance hierarchy, most of the time subordinate flanged males might
have extremely low mating chances and negligible reproductive success (Pradhan et al. 2012).
For subordinate flanged males, the go-and-search strategy could thus represent an alternative
mating strategy. However, compared to unflanged males, they require more energy and incur
in higher costs when moving because of their size. Due to the presence of tigers, a feared
predator, orangutans don’t have the option to move on the ground, a far less energetically
costly form of transportation (Cant 1987).
With the exception of the dominant male, most males in the population remain unflanged
(Dunkel et al 2013). The constant presence of a clearly dominant individual might thus induce
the developmental arrest. The mechanisms that regulate this developmental arrest are, so far,
unknown. Different factors are thought to influence the ‘choice’ of the optimal strategy. It has
been argued that unflanged males use long calls to monitor the presence of flanged males and
only become flanged when the probability of becoming dominant are high (Pradhan et al.
2012). Behavioral observations show that periods of instability in the dominance hierarchy can
indeed promote flanging events (Harrison and Chivers 2007, Utami et al. 2002, Pradhan et al.
2012). It is then possible that within each population the ratio of males pursuing either
strategy reaches an equilibrium in which each males employs the strategy that ensures him
the highest chances of reproducing.
1.5.2 The mating system of the Bornean orangutan (Pongo pygmaeus) Bornean orangutan are more solitary compared to Sumatran ones (Husson et al. 2009).
Females are highly dispersed and only rarely associate with each other. (Singleton et al. 2009).
Due to ecological constraints, the association between males and females is only feasible for
limited amounts of time (Delgado and van Schaik 2000). This situation is substantially different
from what observed in Sumatra. In particular, as a consequence of lower densities and
energetic constraints, monopolization potential for one single male is highly reduced (Utami
12
Atmoko et al. 2009b, Delgado and van Schaik 2000).Field observations suggest that
dominance hierarchy among flanged males is highly unstable and fights are more common
compared to Sumatran populations (Utami Atmoko et al. 2009b). In particular, the presence
of a clearly dominant male is a rare and short-lived event (e.g. in Tuanan) (van Noordwijk,
personal communication). Because prolonged association between one flanged male and the
resident females is deemed to be too energetically costly, several flanged males can have
access to fertile females (Delgado and van Schaik 2000).
Peri-ovulatory females prefer to mate with flanged males (Knott et al. 2010) and estrous
females even approach the long calls of flanged males (Spillman et al. 2010). In contrast,
females with dependent offspring move away from (Spillman et al. 2010). In contrast to what
observed in Sumatra, flanged males also force copulations, similarly to what unflanged males
in both species do (Delgado and van Schaik 2002, Knott et al. 2010).
Because flanged males are always dominant over unflanged males (Mitani 1985), and because
no single male can successfully monopolize the access to females, it has been suggested that
in Bornean populations, unflanged males have a competitive disadvantage (Pradhan et al.
2012). Additionally, due to the lack of predators, flanged males in Borneo can also travel on
the ground and reduce their energetic expenses, making it easier to find females (Cant 1987).
The competitive advantage that flanged males appear to have over unflanged males in Borneo
is one of the possible explanations for the reduced (or lack of) developmental arrest.
Additionally, the perception of dominance instability might promote flanging events (Dunkel
et al. 2013, Pradhan et al. 2012).
1.6 Previous studies of paternity in orangutans In order to further study the evolution and maintenance of alternative mating strategies, an
estimate of reproductive success is required and this can be assessed with an analysis of
paternity. Studies on both Sumatran and Bornean population (Utami et al. 2002, Goossens et
al. 2006) showed that both unflanged and flanged males can successfully sire offspring, thus
confirming that alternative mating strategies can be pursued.
In a Sumatran population at Ketambe, paternities for 10 individuals born over a period of
almost 30 years were assigned using genetic analysis. Four infants were sired by three flanged
males and the remaining six were sired by three unflanged males (Utami et al. 2002). Of the
four infants sired by flanged males, three were sired by two local dominant flanged males and
one by a subordinate flanged male. This suggests that the dominant flanged male has a
reproductive advantage over the subordinate flanged males. Of the six offspring sired by
13
unflanged males, three were sired by a single individual, who later became the local dominant
flanged male. Since the number of flanged males and the number unflanged males were
roughly the same over time, the authors suggest that reproductive success for flanged and
unflanged males is similar, but a few aspects should be considered;
In this population wild-born, ex-captive females were introduced and this might have
influenced the paternity analysis. Two of the three infants that one unflanged male sired,
were conceived with these wild-born, ex-captive females (or daughters of ex-captive), which
could potentially display different sexual behaviors or developmental trajectories from the
ones commonly observed for native females (Utami Atmoko et al. 2009b). At least four of the
six offspring sired by unflanged males were the first offspring for the mother, suggesting that
flanged males prefer more experienced females and might consent unflanged males to mate
with younger, nulliparous females, who also seem to voluntarily interact with unflanged males
(Utami Atmoko et al. 2009b).
A similar study conducted in a Bornean population (Goossens et al. 2006) investigated
paternity of 16 individuals and assigned it to 10. Nine of them were fathered by five flanged
males and one from an unflanged male. Among the flanged males paternity was distributed as
follows: One fathered three infants over a time span of 15 years (1985,1998,2000), a second
male fathered two infants (both in 1990), another male fathered two infants (1990, 1999) and
two more fathered one infant (both in 1990). In contrast with the previously described study
(Utami 2002), in this area unflanged males appear to be far less successful than flanged males.
Even though these studies don’t provide conclusive evidence about the outcome of different
reproductive strategies and the differences between the two orangutan species, they provide
a basis on which I can formulate the hypotheses for my study.
1.7 Paternity assignments in wild animal populations Assignment of paternity in wild populations has always been a difficult process. On one hand,
in most mammal species, maternity can be assigned using behavioral observations, since
pregnancies can be monitored and the offspring is often seen in contact with its mother. A
reliable paternity assignment, on the other hand, requires more than just behavioral data. The
distribution of individuals and the mating history of the mother can help identifying potential
fathers but certainty of paternity is, nowadays, only acquired using genetic data (Jones and
Ardren 2003).
14
The advent of genetic data prompted the development of methods that can be used to assign
paternities. The first methods, used in humans, were based on exclusion criteria (Chakraborty
1974). These methods rely on the genetic incompatibility between offspring and candidate
fathers and aim at excluding as many potential fathers as possible from the pool of candidate
fathers until only one is not excluded. One limitation of this method is that if the number of
potential fathers is high or the power of the markers used not sufficiently high, it’s not always
possible to reduce the list of non-excluded fathers to only one individual.
A second methodology, based on likelihood scores (Thompson 1975, Meagher 1986), aims at
assigning the most likely father to the offspring. For each candidate father, two hypotheses
(H1: candidate is the real father, H0: candidate is not the father) are tested and a likelihood
ratio is calculated. A large likelihood ratio means that H1 is more likely than H0 and therefore
the candidate with the highest ratio is considered to be the true father. Assignment can be
either categorical or fractional; in the first case the offspring is assigned to one potential
father, while in the second case paternity will be distributed proportionally among all the
potential fathers according to the likelihood scores. Whereas the fractional assignment
doesn’t appear to have any biological relevance, it has been observed that for the testing of
particular hypotheses, it can have more statistical power than the categorical approach (eg
Devlin 1988, Smouse and Meagher 1994). Unlike the exclusion approach, a likelihood-based
assignment requires statistical validation and simulated population-specific likelihood ratios
based on allele frequencies allow assignment of paternity within any desired confidence
interval (Marshall et al 1998). Statistical support is needed because the presence of relatives
of the true father or relatives of the offspring (first cousins, half siblings, full-siblings) can lead
to high likelihood scores and thus to false assignments. Tests revealed that the presence of
males that are closely related to the offspring among the potential fathers can indeed cause a
reduction of the confidence of the assignment. Krützen et al. (2004) addressed the problem
using a methodology developed by Brenner and Weir (2003) and incorporated in his analysis a
likelihood ratio for the half-sibling and the true father hypotheses, allowing him to
quantitatively assess whether a parent-offspring relationship was more likely than a half-
siblings one (or vice versa).
Many software packages have been successfully used for paternity assignment in natural
populations (see Jones and Ardren 2003 for a list). For orangutan populations, the software
Cervus 3.0 (Marshall et al 1998, Kalinowsky 2007) has been proven to be a reliable choice
(Utami et al. 2002, Goossens et al. 2006).
15
Another interesting solution is offered in the software COLONY 2.0 (Wang 2012). Instead of
using pairwise comparisons, like most other programs (including Cervus 3.0) do, COLONY 2.0
implements full pedigree likelihood comparisons which test the likelihood of the whole
pedigree rather than just test dyadic relationships. A comparison of the two programs
(Karaket 2012) revealed that COLONY 2.0 required fewer markers to reach the desired level of
accuracy, with the only drawback of necessitating up to three days to perform one single
analysis (opposed to the 30 minutes required by Cervus 3.0).
16
2 HYPOTHESES
The goal of my study is to investigate the reproductive success of male orangutans. Three
main aspects are examined:
1) Paternity Concentration:
In species in which a dominance hierarchy among males is established it is often observed that
the highest ranking individuals sire most of the offspring in their populations (e.g. Bradley
2004, Boesch et al 2006). Among male orangutans strong male-male contest competition for
the access to females is observed and a dominance hierarchy among them is established
(Utami et al 1997). Assessing how paternity is distributed in wild orangutan populations can
help understanding to which extent males can monopolize estrous females.
Paternity concentration can be estimated by assessing the variance of male reproductive
success in the population. In wild populations is not always possible to collect samples from all
the potential fathers and therefore for a portion of the sample it is not possible to successfully
identify the father (e.g. Goossens et al. 2006). In order to overcome this issue and enlarge the
sample size it is possible, by looking at the number of paternal half siblings in a population, to
assess whether offspring for which the fathers wasn’t successfully identified have the same
father or have different ones. In my analysis I will use this approach to try to assess paternity
concentration in my study’s populations. Predictions are presented in table 2-2.
2) Reproductive success of the two male morphs
Both unflanged and flanged males have been proven to successfully sire offspring in wild
orangutan populations (Utami et al. 2002, Goossens et al. 2006) but it is not known yet
whether the two morphs are equally successful or not. I will combine the results from the
paternity analysis with the field observations to try to estimate pro capita reproductive
success values for both morphs. Predictions are presented in table 2-2.
3) Interspecific differences
Based on field observations, differences between the two species might be reflected in their
mating systems and an expected outcome are summarized in Table 2-1.
17
Table 2-1: Comparison of the two orangutan’s species: Relationship between ecology, social- and mating systems. (Adapted from Wich et al (2009) and Delgado and van Schaik (2000))
Sumatra Borneo References
Habitat productivity High quality food all
year round
High quality food is scarce and not always available. Orangutans
have to rely on low quality food.
Marshall 2009
Ecological constraints to association
Virtually absent. High food availability allows
animals to live at higher densities and associate
more often
Association and living at high densities is not
energetically sustainable due to food
scarcity
Husson 2009
Association with local dominant male
Prolonged Short; longer
association is not energetically feasible.
Mitani 1991, Utami 2000
Mating chances for low-ranking flanged
males
Low, females are protected by the dominant male
High, females don’t have protection and
are susceptible to harassment
Utami 2000
Monopolization potential for dominant
flanged male High Low
Delgado and van Schaik 2000
Arrested development Prolonged, affects a high proportion of
males.
Short or even potentially absent.
Dunkel 2013
Occurrence of forced mating
Low, very rarely involves flanged males
High, often involves also flanged males
Miitani 1985, Galdikas 1985
Concentration of paternity
Only a small number of males has access to females. Paternity
concentration is expected to be high
Several males have access to females.
Paternity concentration is
expected to be low
Table 2-2: Expected outcomes of the paternity analysis for the two orangutan species
CONCENTRATION OF PATERNITY SUCCESS OF THE TWO MORPHS
BORNEAN ORANGUTAN
Low, only few paternal half siblings High pro-capita success for flanged males. Low success for unflanged
males
SUMATRAN ORANGUTAN
High, elevated presence of paternal half- and full siblings
High success for the dominant flanged male, very low success for
subordinate flanged males. Unflanged males only successful with nulliparous
females.
18
3 METHODS
3.1 From fecal sample to ready-to-use DNA
3.1.1 Sampling Sampling was conducted in two field sites in Indonesia, Suaq Balimbing and Tuanan, located in
the northern part of Sumatra and Central Kalimantan, respectively. Suaq Balimbing is located
in the western coastal plain of the Leuser and studies of wild orangutans initiated in 1994 but
had to be interrupted in 1999 due to political upheaval. In 2007 researchers have started to
work again in this region. The Tuanan Orang-utan Research Area is located in the south-
eastern part of Kalimantan on Borneo, Indonesia, in peat swamp forest that covers
approximately 7.5 Km2. Studies on wild orangutans have been carried out in this area since
2003.
At both sites, behavioral data have been collected based on a standardized methodology
(http://www.aim.uzh.ch/Research/orangutannetwork/FieldGuidelines.html) and feces
collection for genetic analysis is performed following standard protocols
(http://www.aim.uzh.ch/Research/orangutannetwork/GeneticSamplingProtocol.html). Over
the years, samples were collected in using both the Ethanol-based and the silica-based
methods. All samples were stored at -20°C immediately after arrival in Zürich in other to
prevent further DNA degradation.
3.1.2 Extraction I extracted DNA from fecal samples using the QIAgen mini Stool Kit following the
manufacturer’s instruction with the following modifications:
Incubation time of sample in buffer ASL was modified. For very dry samples, especially silica
dried ones, vortexing for one minute was not enough to ensure optimal homogenization of
the sample with the buffer and therefore increased the incubation time to up to 24 hours.
Elution volume was reduced to 100μl and performed in two steps (2x50 μl) with prolonged
incubation time before both centrifugation steps. After extraction, the extracts were stored in
a fridge until further use.
3.1.3 DNA quantification For this step, I have used a protocol that was developed by Morin et al. (2001) and that has
already been adapted for use with orangutan (Morf, MSc thesis 2009).
19
Table 3-1: Reaction setup for qPCR
Table 3-2: Reaction profile for qPCR
Step: Temperature [°C] Time [min:sec] No. of cycles
Initial Denaturation 95 10:00
Denaturation 95 00:15
55 Annealing 59 00:30
Ct values were estimated by the proprietary software (AB StepOne Sofware v2.3) and a
standard curve was produced using the Ct values of the standards and their initial
concentration. Since the standard curve was then used to estimate the initial DNA
concentration in the extracts, I decided to consider only runs with curves with a correlation
coefficient (R2) higher than 0.95. This ensures a reliable estimate of DNA concentration. I
amplified each extract and standard twice and then averaged the two DNA concentration
estimates.
3.2 Microsatellite genotyping: From DNA to genetic data
For this study I used 28 microsatellite loci (see table 8-1), divided in 5 multiplex setups.
3.2.1 PCR Multiplex
I carried out PCRs in 96-well plates (ABGene SuperPlate 2400) in 8 μl volume. I performed the
PCRs using an AB Verity machine and the reaction profile is described in table 3-4.
Table 3-4 Reaction setup for PCR
Reagent Concentration Final concentration Per reaction [μl]
Template [ml] - - 1
Primer Mix1 10 μM 1 μM 0.8
Multiplex MM 2x 1x 4
ddH2O - - 2.2
Final volume 8 1Primer mixes were prepared independently for each multiplex
Reagent Concentration Final concentration Per reaction [μl]
Template [ml] 2 (5 for standards)
qPCR cmyc F (sigma) 10 μM 0.9 µM 1.8
qPCR cmyc R (sigma) [µM]
10 μM 0.9 µM 1.8
cmyc probe[mM] 10 μM 0.25 µM 0.5
TaqMan Master Mix (AB) 2x
2 1 10
BSA 0.4 μg/μl 0.8
ddH2O 3.1
Final volume 20
20
Table 3-5: Reaction profile for PCR
Step: Temperature [°C] Time [min:sec] No. of cycles
Initial Denaturation 95 15:00 1
Denaturation 94 00:30
45 Annealing 58 01:30
Extension 72 01:00
Final extension 60 30:00 1
I diluted 1 μl of PCR product 20 to 30 times in pure water and then loaded on a 96-well
sequencing plate, along with 10ul of HIDi and 0.07ul of size standard. I denaturated the
double-stranded, diluted PCR products for four minutes at 95°C using an AB Verity machine. I
then sequenced the single-stranded DNA fragments using a 3730xl DNA Analyzer (Applied
Biosystems).
3.2.2 Allele scoring I scored the amplified alleles using the software GeneMapper v4.0 (Applied Biosystem).
GeneMapper v4.0 allows to automatically score alleles but for each reaction, I checked every
call, removed wrongly assigned alleles and added true alleles that were not called by the
software. This evaluation of the quality of the software-made calls was subjective and based
on my personal experience with microsatellite scoring and my familiarity with the loci. I then
exported the called alleles in text file and continued my analysis in Microsoft Excel.
3.2.3 Consensus MACRO This is the final step of the data acquisition process. It is aimed at obtaining a consensus
genotype for each extract. The number of times a genotype has to be observed before
accepting it depends on the initial DNA concentration in the extract (Table 3-6, based on
Morin et al. 2001).
Table 3-6: Conditions for the acceptance of genotypes based on Morin et al. (2001)
3.2.4 Identity analysis After the acquisition of a genotype from a newly analyzed sample I checked whether that
genotype was already present in the database. To do so, I compared the genotype for five loci
DNA Concentration (pg/μl) Heterozygous genotype Homozygous genotype
0-25 Do not use Do not use
25-100 Both alleles appear at least
twice Only one allele, observed at
least 7 times
100-200 Both alleles appear at least
twice Only one allele, observed at
least 4 times
200+ Both alleles appear at least
twice Only one allele, observed at
least 2 times
21
that ensure a high level of allelic diversity and thus a low probability of identity. I performed
this analysis using the software Cervus 3.0 (Marshall et al. 1998). Initially, in order to pick up
genotyping mistakes I have allowed the software to indicate all the matches with up to one
mismatch. When I observed only one mismatch, I re-analyzed the raw data in GeneMapper
v4.0 in order to identify potential genotyping mistakes.
I combined matching genotypes and used them to produce one unique genotype
3.2.5 Usefulness of genetic markers Error rates can be estimated using repeated genotyping of randomly chosen individuals by
dividing the number of mismatches for the total number of comparisons (Bonin et al 2004).
This approach allows to estimate overall error rates but also locus-specific error rates. I tested
each locus for Hardy-Weinberg disequilibrium and each locus pair for Linkage disequilibrium. I
carried out this analysis in Genepop v4 (Rousset 2008) and Genalex 6.5 (Peakall and Smouse
2012). After the preliminary tests for Hardy-Weinberg disequilibrium and linkage
disequilibrium, for each population I estimated several descriptive parameters using Cervus
3.0 (Marshall 1998, Kalinowski 2007). These parameters offer an overview of the markers
used in the study and allow to estimate the power of the dataset. In particular, the number of
allele per locus (NA), expected (HE) and observed (HO) heterozygosity, Polymorphic
Information Content (PIC) and probability of identity for both unrelated individuals (Pi) and
siblings (Pis) are measured in Cervus 3.0.
In order to further estimate the suitability of the genetic markers, I calculated the Power for
relationship inference (PWR) using KinInfor 1.0 (Wang 2006). PWR indicates the likelihood of
the set of loci to discriminate between two pedigree relationships and it is a reliable
measurement for the assessment of multilocus informativeness (Wang J, personal
communication). For my analysis I tested the ability of my dataset to discriminate between
Parent-offspring/unrelated, half-sibling/unrelated, half-sibling/first cousin and Parent-
offspring-Full sibling.
3.2.6 Parentage analysis I assigned paternities using the software Cervus 3.0 and COLONY 2.0. All males in the dataset
were included as potential fathers with the exclusion of pre-dispersal males. Paternities were
assigned at the 80% confidence level. An important parameter for paternity analysis is the
proportion of sampled fathers and this value is particularly difficult to estimate in orangutans
because of the size of the study sites and the unpredictable ranging patterns of males. The
percentage of sampled males was estimated using the number of genotyped males and the
22
number of males observed in the field (Dunkel et al 2013). I set this value at 25% for both
populations.
I carried out a maternity analysis for all individuals in the dataset. I included every adult
female in the pool of potential mothers and the analysis was performed using the software
Cervus 3.0 and COLONY 2.0. For each candidate offspring, females that are known to be
younger than the offspring were excluded from the list of candidate mothers. This step,
although not strictly necessary, can increase the power of the analysis in presence of highly
related individuals. Maternities were accepted at the 80% confidence level. The proportion of
mothers sampled was estimated based on field observations and on the number of individuals
sampled.
3.2.7 Developmental status of the father and reproductive history of the mother Developmental status of the males were assigned using field observations (information
provided by Carel van Schaik and Maria van Noordwijk). In order to assess the reproductive
history of the mother I have used the results of my maternity analysis and field observations.
In some instances I could not assess with certainty whether the mother was nulliparous or
multiparous because, due to strong male dispersal, I have almost only identified female
offspring and have no way of identifying previous male offspring that females could have had.
3.2.8 Estimate of paternity concentration In order to estimate the number of fathers present in the study populations, I used the
software COLONY 2.0 (Wang 2012). Even in absence of the real fathers, the software can
calculate the confidence for the assignment of dyads of full- or half-siblings. This procedure
will be used to estimate paternity concentration by assessing the number of successful males
at any given time without requiring the sampling of all the true fathers. I accepted half-sib pair
that were assigned at a confidence level of 80%. After assessing the number of fathers in the
population I estimated the level and significance of reproductive skew among males. I have
done that only for Suaq and for the time period 1988-1997. I calculated the B-index (Nonacs
2000) using Skew Calculator 2013 (Nonacs 2003). The B-index estimates the strength and
significance of reproductive skew as follows:
B<0 Reproductive success is equally shared
B=0 Reproductive success is randomly distributed among males
B>0 Reproductive success is skewed and not equally shared
For the data regarding male numbers and presence, I have used the data present in Dunkel et
al. (2013). Presence estimates were divided in three categories: 50%, 25% and 5%. In order to
23
obtain a conservative estimate of paternity concentration, unknown fathers were assumed to
belong to the 25% presence category.
4 RESULTS
4.1 Suaq Paternity analysis in Cervus 3.0 revealed 10 father-offspring pairs (Table 4-1). 9 out of 10
paternities were assigned to offspring of known mother. In all cases, the confidence for the
mother-father-offspring trio reached 95%. Paternities were assigned to 6 different males.
Arno obtainined 4 paternities, Ria obtained 2, Mack, Bestel, Dian and Sumbing each obtained
1 paternity.
Table 4-1: Paternity assignments with Cervus 3.0 for Suaq (LC= Number of loci compared, LMM= Number of mismatches, confidence: *=95%, +=80%)
Offspring ID
Mother ID
Father ID Lc LMM Pair
ΔLOD
Pair confidence
Trio
ΔLOD
Trio confidence
ROBBI LENA ARNO 22 0 10.8 * 13.5 *
ELLIE FRISKA ARNO 22 0 4.59 + 9.41 *
LISA CISSY ARNO 22 0 4.69 + 13.1 *
TINA RAFFI ARNO 22 0 9.53 * 15.3 *
NIBLA INTAY BESTEL 21 0 7.88 * 11.1 *
DODI CHICK DIAN 21 0 6.13 * 13.6 *
UNO BIB MACK 21 1 2.45 + 6.75 *
CHINDY CISSY RIA 23 1 5.23 * 10.8 *
LILLY LISA RIA 23 0 7.36 * 10.0 *
PRECILLA na SUMBING 22 0 5.62 * na na
Paternity analysis in COLONY confirmed the paternity assignments obtained with Cervus 3.0.
All but one assignments reached the 95% and one assignment reached and the 80%
confidence level. No paternal half-sib pair were identified at the 80% significance level (table
4-2). Table 4-2 also includes the information about the year of conception, father’s
developmental stage (unflanged or flanged), father’s rank (dominant or subordinate) and the
mother’s reproductive history (nulliparous vs multiparous).
24
Table 4-2 Paternity assignments and Half-sibship assignments for Suaq with COLONY 2.0
Offspring ID
Mother ID
Father ID Confidence level
Year of conception
Father’s status at
conception
Father’s Rank
Mother’s reproductive
history
LILLY LISA RIA 1.000 (*) 2001 Unknown UNK NP
UNO BIB MACK 1.000 (*) 1988 Flanged UNK UNK
LISA CISSY ARNO 1.000 (*) 1988 Flanged D UNK
TINA RAFFI ARNO 1.000 (*) 1996 Flanged D NP
NIBLA INTAY BESTEL 1.000 (*) 1996 Unflanged S NP
ROBBI LENA ARNO 1.000 (*) 1996 Flanged D NP
DODI CHICK DIAN 1.000 (*) 1991-92 Unknown S UNK
CHINDY CISSY RIA 0.997 (*) 2002 Unknown UNK MP
ELLIE FRISKA ARNO 0.986 (*) 1998 Flanged D MP
PRECILLA UNKN_1 SUMBING 0.905 (+) Unknown Unknown UNK -
SHERA CHICK UNKN_1 1997 - - MP
FREDDY FRISKA UNKN_2 Half-sib: 0.703 (-)
2004 - - MP
BARBARA UNKN_2 UNKN_2 Unknown - - -
ABBY UNKN_2 UNKN_3 Half-sib: 0.117 (-)
Unknown - - -
HERDY FRISKA UNKN_3 1988 - - MP
HALTE FRISKA UNKN_4 Unknown - - UNK
TEDI HALTE UNKN_5 1996 - - MP
LENA DIANA UNKN_6 1982 - - -
DIDDY DODI UNKN_7 2005 - - MP
NUK - UNKN_8 2006 - - -
GANI UNKN_3 UNKN_9 Unknown - - -
RAFFI HALTE UNKN_10 Unknown - - UNK
The B-index for the period 1988-1997 was positive and estimated at 0.0464, indicating
significant (p=0.048) reproductive skew (see Table 8-8).
4.2 Tuanan Paternity analysis in Cervus 3.0 revealed 8 father-offspring pairs (Table 4-5). Paternities
assigned to individuals with known mothers reached the 95% for the trio confidence level,
while one paternity for an individual without known maternal genotype reached 80%
confidence level. Paternities were assigned to 4 males, Kentung, Guapo, Niko and Wodan.
Table 4-3: Paternity assignments with Cervus 3.0 for Tuanan (LC= Number of loci compared, LMM= Number of mismatches, confidence: *=95%, +=80%)
Offspring ID
Mother ID
Father ID Lc LMM Pair
ΔLOD Pair confidence
Trio
ΔLOD
Trio confidence
IDO INUL GUAPO 10 0 4.73 * 8.17 *
JERY JINAK NIKO 18 0 6.80 * 13.2 *
MILO MINDY NIKO 20 0 8.90 * 13.7 *
STREISEL SIDONY WODAN 25 0 10.9 * 14.4 *
PUMUCKL PINKY GUAPO 10 1 1.40 + 2.51 *
LOLO SIDONY KENTUNG 23 0 9.48 * 13.1 *
NANIO - KENTUNG 23 2 2.76 + - -
TIKUS TALIA KENTUNG 25 1 10.2 * 16.4 *
25
A paternity analysis in COLONY confirmed the paternities assigned with Cervus 3.0. One
paternal half-sib pair was identified at the 95% confidence level (table 4-4). Table 4-4 also
includes the information about the year of conception, father’s developmental stage
(unflanged or flanged), father’s rank (dominant or subordinate) and the mother’s reproductive
history (nulliparous vs multiparous). Four paternities were assigned to flanged males and one
to an unflanged male. In three instances, I could not reliably assess the developmental status
of the father.
Table 4-4: Paternity assignments for Tuanan with COLONY 2.0
Offspring ID
Mother ID
Father ID Confidence level
Year of conception
Father’s status at
conception
Father’s Rank
Mother’s reproductive
history
JERY JINAK NIKO 1.000 (*) 2002 Flanged ? MP
MILO MINDY NIKO 1.000 (*) 2001 Flanged ? Unknown
STREISEL SIDONY WODAN 1.000 (*) 2002 Unflanged ? MP
IDO INUL GUAPO 0.898 (+) 2001 Flanged ? MP
PUMUCKL PINKY GUAPO 0.896 (+) 2001 Flanged ? Unknown
LOLO SIDONY KENTUNG 1.000 (*) ~1995 Unknown ? Unknown
NANIO - KENTUNG 0.997 (*) 1997 Unknown ? Unknown
TIKUS TALIA KENTUNG 1.000 (*) 1999 Unknown ? Unknown
CIKIPOS WILMA UNKN_1 Full-sib: 0.574 (-)
Unknown - - Unknown
FRODO WILMA UNKN_1 1997 - - Unknown
DESY INUL UNKN_2 - Unknown - - Unknown
JIP JUNI UNKN_3 - 2006 - - NP
JUNI JINAK UNKN_4 - 1994 - - MP
KERRY JINAK UNKN_5 - Unknown - - Unknown
KONDOR KERRY UNKN_6 - 1999 - - Unknown
MINDY JINAK UNKN_7 Half-sib: 0.963 (*)
Unknown - - Unknown
WILMA UNKN_1 UNKN_7 Unknown - - -
SUSI WILMA UNKN_8 - 2003 - - MP
TALIA - UNKN_9 - Unknown - - -
26
5 DISCUSSION In my thesis, I explored the patterns of male reproductive success in a population of Sumatran
orangutans (P. abelii) and one of Bornean orangutans (P. pygmaeus wurmbii). I assigned 10
paternities in the Sumatran population and 8 in the Bornean one. Additionally, I have
managed to provide evidence for high paternity concentration for the Sumatran population.
5.1 Reproductive success of male orangutans
5.1.1 Suaq In Suaq a flanged male, Arno, sired four of six sampled offspring that were born between the
late 1980s and the late 1990s. He was reported to be the undisputed dominant male between
1994 (when the observations started) and 1999 (Singleton and van Schaik 2002). I confirmed
that at least three of the offspring born around 1996 were indeed sired by Arno. The fact that
he also sired at least one of the six offspring born in 1988 suggests that his tenure might have
started in the late 1980s.
Between 1996 and 1997, at least 9 offspring were born and Arno sired three of the five infants
analyzed in this study. This suggests that paternity concentration might be high, reflecting a
high monopolization potential. Based on field observations, this result is not surprising,
because Arno was consistently seen in consortship with females. The calculation of the B-
index (Nonacs 2003) confirms that male reproductive success in Suaq for the period 1988-
1997 is significantly skewed and concentrated in the local dominant male.
The paternity obtained by Mack in 1988 should not be too surprising as he was one of the only
three flanged males who regularly ranged within the study area (Singleton and van Schaik
2000). It is even possible that in the late 80s he was the local dominant male.
In 1996, Bestel was unflanged and his success with a young, nulliparous female (Intay) is
consistent with what observed in another population of Sumatran orangutans. Schürmann
(1982) noted in Ketambe that unflanged males were the first ones to show interest for a
young female. Whereas she showed a clear interest in the local dominant male, he did not
reciprocate and started to be interested in her only the year before she conceived
(Schürmann 1982). It is then possible that if females get pregnant when still in the phase of
voluntary courtship with unflanged males, the paternity of the offspring might be acquired by
one of these unflanged males (Utami Atmoko 2009b).
Around the end of 1999 Arno was challenged by several immigrant flanged males who
entered the area following extensive logging in neighboring forest patches (van Schaik,
27
personal communication). In 2001 and 2002, paternity of two offspring was assigned to Ria, a
male who up to 1999 was unflanged. Ria’s success with one of the older resident females,
Cissy, is a clear sign of the weakening of Arno’s dominance (or his disappearance). Cissy and
Arno were in fact seen to be often in association throughout the 1990s and already had a
daughter together in 1988.
Subordinate flanged males and unflanged males appear to be equally (un)successful and are
significantly less successful than the dominant flanged male. When the dominance hierarchy is
stable (most of the time), unflanged males might be more successful than subordinate flanged
males. Overall, flanged males (dominant and subordinate) are more successful than unflanged
males. As unflanged males represent the majority of males in the area (Dunkel et al 2013), I
suggest that their pro capita reproductive success is low, despite the high levels of sexual
activity observed (Utami Atmoko et al. 2009b).
My results are partially discordant with the findings for another Sumatran population,
Ketambe, in which the dominant males lost several paternities to unflanged males. In that
case, one male dominated for approximately 18 years but only obtained one paternity out of
four that were investigated when he was present, losing three to two unflanged males (Utami
et al. 2002). The differences between the two results could be caused by the sexual behavior
of the several nulliparous females. These females descend from a line of ex-rehabilitant
females and were likely to have developed earlier do to the provisioning of high quality food
from humans (Utami Atmoko et al. 2009). These females might have gotten pregnant in the
phase of voluntary courtship with unflanged males.
5.1.2 Tuanan For Tuanan, 8 paternities out 18 investigated offspring were assigned. One male of unknown
developmental status obtained three paternities, two males (likely to be flanged at the time of
conception) sired two offspring and one paternity was assigned to an unflanged male. The
three offspring sired by Kentung were born in the mid- and late nineties, while the remaining
five offspring were all born between 2001 and 2003, supporting the idea that in this
population several males are mating with receptive females.
The relatively low rate of paternity assignment might be a reflection of the high number of
transient, non-local males observed in the field. Flanged males come and go and only rarely
“settle” in the study area (Dunkel et al 2013). Even though one of the frequently present
males, Niko, was successful at least twice, it seems that also several transient males might be
reproductively successful.
28
The success for one unflanged male suggests that these males are copulating with estrous
females. In another Bornean site, Gunung Palung, the majority of copulations for
periovulatory females involved flanged males, but a copulation between a periovulatory
female and an unflanged male was observed (Knott et al. 2010). In this case the copulation
was forced and this observation is compatible with the go-and-search strategy that unflanged
males are believed to employ in Borneo.
Based on my results, the hypothesis that in P. pygmaeus populations several flanged males
are successful whereas unflanged males are only rarely successful cannot be rejected, but
more data is needed. An estimate of reproductive skew was not possible to achieve. Most
offspring in my sample were born before research started in Tuanan and thus information
about male presence is lacking.
5.2 How do these results improve our understanding of orangutan’s
mating system?
5.2.1 Interspecific differences My results suggest that the levels of paternity concentration in the two species differ. As
predicted, the dominant male in Suaq acquired the majority of the paternities during his
tenure, whereas in Tuanan several flanged males are successful at the same time. However,
Kentung had high reproductive success in a 5 year window, suggesting that some individuals
can be more successful and that paternity may only be concentrated in a few individuals.
Unflanged males are successful in both populations and this result is not surprising for the
Sumatran population. In contrast, unflanged males in Borneo were thought to be very unlikely
to mate with receptive females (Knott et al. 2010) and the success of a Bornean unflanged
male raises new questions about their reproductive strategies.
5.2.2 Reproductive success of the unflanged male Both mine and previous results (Utami 2002), suggest that Sumatran unflanged males might
be particularly successful with young nulliparous females. In primate species, male choice for
older females might be adaptive (Anderson 1986). Extensive offspring care is required and
more experienced females have higher chances of successfully raising an offspring (Paul 1998,
Anderson 1986). Additionally, nulliparous females might require more cycles to get pregnant
(Gesquiere et al. 2007).
In chimpanzee (Pan troglodytes), the top ranking males have priority of access to females, and
subordinate males may attempt to achieve paternities by mating opportunistically. Each low
ranking male can sire between 0% and 10% of the offspring in the group. Younger, low-
29
ranking males only sire the offspring of younger females (Wroblewski 2009, Muller et al.
2007).
In a multi-male mountain gorilla (Gorilla beringei) group, subordinate silverbacks were found
to be more likely to copulate with nulliparous females than the alpha male. In this case, all the
nulliparous females were known (or very likely) to be the daughters of the alpha silverback
(Stoinski et al. 2009). Another study on mountain gorilla, the subordinate males achieve
approximately 15% of the paternities and in this case it was suggested that their success is
caused by a lack of control by the dominant silverback (Bradley et al. 2005).
In baboons (Papio cynocephalus), adolescent females conceive in only 1% of the cycles (in
contrast to 30% for adult females). Alpha males also in this species mate preferentially with
older females (Gesquiere et al. 2007). In rhesus macaques (Macaca mulatta), adolescent
females present sexually more often to low-rank males. Compared to adult females,
adolescents also engage more frequently in copulation with low-rank males and extra-group
males (Perry and Manson 1995).
These observations might help understanding the pattern seen in Sumatran orangutans. The
dominant flanged male often ignores the proceptive behavior of adolescent females (Utami
Atmoko et al. 2009b).As the tenure of a dominant male is unlikely to reach 14 years, the age
at which potential daughters become sexually active, I suggest that is unlikely that flanged
males are not interested in young females to avoid inbreeding. Field observations suggest that
nulliparous females require remarkably long periods (months, or even years) of sexual activity
to get pregnant (Carel van Schaik, personal communication), indicating that their fertility
might be initially much lower compared to older females. I suggest that the dominant male
focuses his reproductive efforts on older females and only begins to show interest in
nulliparous females when the chances of fertilization outweigh the effort. This strategy of the
dominant male opens a window of opportunity to unflanged males. Even though most of the
copulations with these females won’t lead to conception, it is possible that every now and
then a copulation between an unflanged males and a nulliparous female will result in a
pregnancy.
5.2.3 Evidence for infanticide in Sumatran orangutans? In primate species infanticide is a widespread behavior (van Schaik and Janson 2000, Watts
1989, Swedell and Tesfaye 2003, Struhsaker 1977, Clarke et al. 1994) and the potential
benefits for the male perpetrators are of various nature (Ebensperger 1998).
30
The disappearance of at least four offspring in Suaq at the end of the 1990s has been
observed (van Noordwijk and van Schaik 2005). As infant mortality for the first year in this
species is generally very low at around 8% (Wich et al.2004), the disappearance of more than
30% of the infants in this population has been linked to human disturbance (logging and pet
trade). In most cases involving the capturing of an offspring, the mothers are killed (Nijman
2005). However, in Suaq the mothers of the deceased offspring were still alive after the loss of
their offspring. This suggests that infanticide could be a more reasonable explanation for the
high mortality rates.
In orangutans, infanticide has never been documented (Beaudrot et al. 2009), even though
there are reasons to believe that for orangutan males it would be relatively to kill unweaned
offspring (van Schaik and Kappeler 1997). It is believed that Female orangutans of both
species mate promiscuously and this behavior seems to induce paternity confusion, an
effective counterstrategy to infanticide (Knott 2010, Stumpf et al. 2008).
Beaudrot et al. (2009) identified three conditions that need to be met in order to make
infanticide an effective strategy for males: First, for males infanticide is only adaptive when
the probability of siring the female’s next offspring is higher than the probability of having
sired the current offspring. Second, infanticide is only adaptive if females resume ovarian
activity quickly after losing the current offspring. Third, males need to have access to the
female after killing her current offspring.
The authors argue that at least two of these conditions are not met in Sumatran orangutans.
In particular, based on a previous study of paternity for Ketambe (Utami et al. 2002), they
argue that resident subordinate flanged males can be reproductively successful and thus
would not benefit from infanticide. My results for Suaq seems to paint a different picture.
Subordinate flanged males obtain virtually no copulations (Singleton and van Schaik 2002) and
the results of my paternity analysis suggest that paternity is highly concentrated in the
dominant male. I therefore argue that for both resident and transient subordinate flanged
males, infanticide might be advantageous.
Beaudrot et al. (2009) also argue that female orangutans might not resume ovarian activity
after the loss of an offspring. My maternity analysis confirms that Friska, who had lost a one
year old offspring in 1998, had a new baby around 1999-2000. As energy availability has been
linked to fertility in orangutans (Knott et al. 1999, but also Wich et al. 2006), it is possible that
females in the energy-rich forests of Suaq might have the ability to give birth to a new
offspring soon after losing one.
31
Thirdly, Beaudrot et al. (2009) argued that males in Sumatra should not kill infants because
the access to the mother after the death is not guaranteed. This might be true most of the
time but in the rare event of a male taking over a population, his future access to females is
granted (Utami and Mitra Setia 1995).
The situation observed in Suaq at the end of the 1990s is peculiar. New, unknown flanged
males were pushed into the study area by human activity and deposed the dominant male. In
this situation it seems that it would be advantageous for the new males to kill the current
offspring and might explain why infant mortality at the end of the 1997 and 1998 was so high.
In conclusion, a combination of high paternity concentration and monopolization potential,
high infant mortality and dominance takeover by transient males could have led to infanticide
in Suaq.
5.3 Limitations of this study and outlook The lack of knowledge about male reproductive success does not lie in the lack of effort from
the scientific community but are certainly linked to various aspects of orangutan’s research.
Firstly, the collection, storage and handling of non-invasive fecal samples in the Indonesian
field sites represent an obstacle. Secondly, the effort required to obtain reliable genotypes
from these samples is less time- and cost-effective compared to other types of samples
(blood, tissue) and might discourage scientist from working with them. Thirdly, due to
unpredictable ranging patterns and low densities, individuals are harder to find and to study
over a long period. Lastly, orangutans have the slowest life history among non-human
primates and females only produce offspring every 6-9 years (Galdikas and Ashbury 2012,
Wich et al. 2004) making the gathering of sufficient data for paternity analysis very time
consuming.
For my study, estimated year of birth, male developmental status at the time of conception
and female reproductive history were generally very difficult to assess despite the longevity of
the research projects for both populations. For this reason, in most cases reliable information
for one or more attributes was missing and strongly affected my ability to determine the
reproductive success of the two male morphs and how it is linked with the mother’s
reproductive history.
For what concerns the genetic analysis, I have used two powerful microsatellite datasets
which allow to successfully discriminate between individuals and assign paternities at a
satisfying confidence level (Fig 8-4- and 8-5). I have nonetheless encountered some
32
limitations. As revealed from my analysis, intricate pedigree relationships were difficult to
assess. The assessment of paternal half-sibs was very difficult, even in presence of the
maternal genotypes. I suggest that the use of more markers might increase the power for
relationship inference and more extensive sampling might provide better estimates of allele
frequencies.
However, for the parentage analysis I have managed to achieve the 95% confidence level even
in presence of mismatches and with an error rate set at 1%. This suggests that my dataset is
powerful enough for likelihood analysis. In fact, even in presence of a mismatch, likely caused
by genotyping mistake or a mutation (e.g. Nsubuga 2008), the remaining loci offer strong
enough evidence in favor of a true parent-offspring relationship.
Despite the difficulties encountered, I think that the two datasets have the potential to
unravel the mating systems for both species. For Suaq, field observation for the last 6 years
suggest that a stable dominance hierarchy has been once again established. Paternity for the
yet-to-be sampled 13 offspring born between 2003 and 2014 could reveal a similar picture to
what was observed in the 1990s with Arno.
Similarly, in Tuanan, new samples have been collected and paternity for 17 more offspring
born between 2003 and 2014 could in the future be assessed. The likelihood of having
sampled the father of these recently born offspring is higher. In my data, most of the offspring
were born before research started in and the father was likely gone by the time sample
collection started in 2003. I have assigned paternity to five out of seven offspring analyzed
who were born after the year 2000, and assigned 3 out of 10 analyzed who were born before
the year 2000. Knowing that my markers are powerful enough to identify fathers, I suggest
that in those cases the true fathers have not been sampled. Given the rather unstable social
organization of flanged males in Borneo (Dunkel et al 2013), I suggest that the sampling of as
many males as possible around time of conception is necessary for paternity analysis. The
assignment of more paternities, in combination with the extensive behavioral, ecological and
physiological data collected in the past ten years in Tuanan, might finally shed some light on
this two species’ mating systems.
33
6 AKNOWLEDGMENTS I would like to thank Michael Krützen for giving me the chance to work in his group and for the
support provided in the past year. I want to thank Maria van Noordwijk, Carel van Schaik,
Caroline Schuppli and Brigitte Spillman for providing the information about the orangutans in
the field. I also want to thank the members of the evolutionary genetics group for the help in
the lab and the moral support, as well as the A.H. Schultz Stiftung for the greatly appreciated
financial support.
7 LITERATURE CITED Altmann, S. A. 1962. A field study of the sociobiology of rhesus monkeys, macaca mulatta*. Annals of the New York Academy of Sciences, 102, 338–435.
Anderson, C. M. 1986. Female age: Male preference and reproductive success in primates. International Journal of Primatology, 7, 305–326.
Anderson, M. J. & Dixson, A. F. 2002. Sperm competition: motility and the midpiece in primates. Nature, 416, 496–496.
Arora, N., Van Noordwijk, M., Ackermann, C., Willems, E., Nater, A., Greminger, M., Nietlisbach, P., Dunkel, L., Utami Atmoko, S., Pamungkas, J. & others. 2012. Parentage-based pedigree reconstruction reveals female matrilineal clusters and male-biased dispersal in nongregarious Asian great apes, the Bornean orang-utans (Pongo pygmaeus). Molecular Ecology,
Beaudrot, L. H., Kahlenberg, S. M. & Marshall, A. J. 2009. Why male orangutans do not kill infants. Behavioral ecology and sociobiology, 63, 1549–1562.
Berard, J. 1999. A four-year study of the association between male dominance rank, residency status, and reproductive activity in rhesus macaques (Macaca mulatta). Primates, 40, 159–175.
Boesch, C., Kohou, G., Néné, H. & Vigilant, L. 2006. Male competition and paternity in wild chimpanzees of the Ta"\i forest. American Journal of Physical Anthropology, 130, 103–115.
Borries, C., Savini, T. & Koenig, A. 2011. Social monogamy and the threat of infanticide in larger mammals. Behavioral ecology and sociobiology, 65, 685–693.
34
Bradley, B. J., Doran-Sheehy, D. M., Lukas, D., Boesch, C. & Vigilant, L. 2004. Dispersed male networks in western gorillas. Current Biology, 14, 510–513.
Bradley, B. J., Robbins, M. M., Williamson, E. A., Steklis, H. D., Steklis, N. G., Eckhardt, N., Boesch, C. & Vigilant, L. 2005. Mountain gorilla tug-of-war: silverbacks have limited control over reproduction in multimale groups. Proceedings of the National Academy of Sciences of the United States of America, 102, 9418–9423.
Brenner, C. & Weir, B. 2003. Issues and strategies in the DNA identification of World Trade Center victims. Theoretical population biology, 63, 173–178.
Cant, J. G. 1987. Positional behavior of female Bornean orangutans (Pongo pygmaeus). American Journal of Primatology, 12, 71–90.
Chakraborty, R., Shaw, M. & Schull, W. J. 1974. Exclusion of paternity: the current state of the art. American Journal of Human Genetics, 26, 477.
Clarke, M. R., Zucker, E. L. & Glander, K. E. 1994. Group takeover by a natal male howling monkey (Alouatta palliata) and associated disappearance and injuries of immatures. Primates, 35, 435–442.
Clutton-Brock, T. H. & Parker, G. A. 1992. Potential reproductive rates and the operation of sexual selection. Quarterly Review of Biology, 437–456.
De Waal, F. B. M. 1995. Bonobo sex and society. Scientific American, 272, 82–88.
Delgado, R. A. 2003. The function of adult male long calls in wild orangutans (Pongo pygmaeus).
Delgado, R. A., Lameira, A., Davila Ross, M., Husson, S. J., Morrogh-Bernard, H. C. & Wich, S. A. 2009. Geographical variation in orangutan long calls. Orangutans: geographic variation in behavioral ecology and conservation. Oxford University Press, Oxford,
Delgado Jr, R. A. & Van Schaik, C. P. 2000. The behavioral ecology and conservation of the orangutan (Pongo pygmaeus): a tale of two islands. Evolutionary Anthropology: Issues, News, and Reviews, 9, 201–218.
Devlin, B., Roeder, K. & Ellstrand, N. 1988. Fractional paternity assignment: theoretical development and comparison to other methods. Theoretical and Applied Genetics, 76, 369–380.
Dixson, A., Bossi, T. & Wickings, E. 1993. Male dominance and genetically determined reproductive success in the mandrill (Mandrillus sphinx). Primates, 34, 525–532.
35
Doran, D. M. & Mcneilage, A. 1998. Gorilla ecology and behavior. Evolutionary Anthropology: Issues, News, and Reviews, 6, 120–131.
Dunkel, L. P., Arora, N., van Noordwijk, M. A., Atmoko, S. S., Putra, A. P., Krützen, M. & van Schaik, C. P. 2013. Variation in developmental arrest among male orangutans: a comparison between a Sumatran and a Bornean population. Frontiers in Zoology, 10, 12.
Ebensperger, L. A. 1998. Strategies and counterstrategies to infanticide in mammals. Biological Reviews, 73, 321–346.
Emlen, S. T. & Oring, L. W. 1977. Ecology, sexual selection, and the evolution of mating systems. Science, 197, 215–223.
Galdikas, B. M. F. 1985. Orangutan sociality at Tanjung Puting. American Journal of Primatology, 9, 101–119.
Gerloff, U., Hartung, B., Fruth, B., Hohmann, G. & Tautz, D. 1999. Intracommunity relationships, dispersal pattern and paternity success in a wild living community of bonobos (Pan paniscus) determined from DNA analysis of faecal samples. Proceedings of the Royal Society of London. Series B: Biological Sciences, 266, 1189–1195.
Gesquiere, L. R., Wango, E. O., Alberts, S. C. & Altmann, J. 2007. Mechanisms of sexual selection: sexual swellings and estrogen concentrations as fertility indicators and cues for male consort decisions in wild baboons. Hormones and Behavior, 51, 114–125.
Goodall, J. 1986. The chimpanzees of Gombe: Patterns of behavior.
Harrison, M. E. & Chivers, D. J. 2007. The orang-utan mating system and the unflanged male: A product of increased food stress during the late Miocene and Pliocene? Journal of human evolution, 52, 275–293.
Higham, J. P. & Maestripieri, D. 2010. Revolutionary coalitions in male rhesus macaques. Behaviour, 147, 13–14.
Husson, S. J., Wich, S. A., Marshall, A. J., Dennis, R. D., Ancrenaz, M., Brassey, R., Gumal, M., Hearn, A. J., Meijaard, E., Simorangkir, T. & others. 2009. Orangutan distribution, density, abundance and impacts of disturbance. Orangutans: Geographic variation in behavioral ecology and conservation, 77–96.
Isbell, L. A. 1994. Predation on primates: ecological patterns and evolutionary consequences. Evolutionary Anthropology: Issues, News, and Reviews, 3, 61–71.
36
Janson, C. H. & Goldsmith, M. L. 1995. Predicting group size in primates: foraging costs and predation risks. Behavioral Ecology, 6, 326–336.
Jones, A. G. & Ardren, W. R. 2003. Methods of parentage analysis in natural populations. Molecular Ecology, 12, 2511–2523.
Kalinowski, S. T., Taper, M. L. & Marshall, T. C. 2007. Revising how the computer program CERVUS accommodates genotyping error increases success in paternity assignment. Molecular ecology, 16, 1099–1106.
Kappeler, P. M. & van Schaik, C. P. 2002. Evolution of primate social systems. International journal of primatology, 23, 707–740.
Knott, C. D. 1999. Reproductive, physiological and behavioral responses of orangutans in Borneo to fluctuations in food availability.
Knott, C. D., Thompson, M. E., Stumpf, R. M. & McIntyre, M. H. 2010. Female reproductive strategies in orangutans, evidence for female choice and counterstrategies to infanticide in a species with frequent sexual coercion. Proceedings of the Royal Society B: Biological Sciences, 277, 105–113.
Krützen, M., Barré, L. M., Connor, R. C., Mann, J. & Sherwin, W. B. 2004. “O father: where art thou?”—Paternity assessment in an open fission-fusion society of wild bottlenose dolphins (Tursiops sp.) in Shark Bay, Western Australia. Molecular Ecology, 13, 1975–1990.
Kutsukake, N. & Nunn, C. L. 2006. Comparative tests of reproductive skew in male primates: the roles of demographic factors and incomplete control. Behavioral Ecology and Sociobiology, 60, 695–706.
Kvarnemo, C. & Ahnesjo, I. 1996. The dynamics of operational sex ratios and competition for mates. Trends in Ecology \& Evolution, 11, 404–408.
Lawson Handley, L. & Perrin, N. 2007. Advances in our understanding of mammalian sex-biased dispersal. Molecular Ecology, 16, 1559–1578.
Lehmann, J. & Boesch, C. 2004. To fission or to fusion: effects of community size on wild chimpanzee (Pan troglodytes verus) social organisation. Behavioral Ecology and Sociobiology, 56, 207–216.
Lukas, K. E., Kuhar, C. W. & Stoinski, T. S. 2010. Behavioral response of captive western lowland gorillas (Gorilla gorilla gorilla) to the death of silverbacks in multi-male groups. Zoo biology, 29, 16–29.
37
MacKinnon, J. 1974. The behaviour and ecology of wild orang-utans (Pongo pygmaeus). Animal Behaviour, 22, 3–74.
Marshall, T., Slate, J., Kruuk, L. & Pemberton, J. 1998. Statistical confidence for likelihood-based paternity inference in natural populations. Molecular Ecology, 7, 639–655.
Marshall, A. J., Ancrenaz, M., Brearley, F. Q., Fredriksson, G. M., Ghaffar, N., Heydon, M., Husson, S. J., Leighton, M., McConkey, K. R., Morrogh-Bernard, H. C. & others. 2009. The effects of forest phenology and floristics on populations of Bornean and Sumatran orangutans. Orangutans: geographic variation in behavioral ecology and conservation. Oxford: Oxford University Press. p, 97–117.
Mass, V., Heistermann, M. & Kappeler, P. 2009. Mate-guarding as a male reproductive tactic in Propithecus verreauxi. International journal of primatology, 30, 389–409.
Massen, J. J., Overduin-de Vries, A. M., de Vos-Rouweler, A. J., Spruijt, B. M., Doxiadis, G. G. & Sterck, E. H. 2012. Male mating tactics in captive rhesus macaques (Macaca mulatta): the influence of dominance, markets, and relationship quality. International journal of primatology, 33, 73–92.
Matsubara, M. 2003. Costs of mate guarding and opportunistic mating among wild male Japanese macaques. International Journal of Primatology, 24, 1057–1075.
Matsumoto-Oda, A. 1999. Female choice in the opportunistic mating of wild chimpanzees (Pan troglodytes schweinfurthii) at Mahale. Behavioral Ecology and Sociobiology, 46, 258–266.
Meagher, T. R. 1986. Analysis of paternity within a natural population of Chamaelirium luteum. 1. Identification of most-likely male parents. American Naturalist, 199–215.
Mitani, J. C. 1985. Mating behaviour of male orangutans in the Kutai Game Reserve, Indonesia. Animal Behaviour, 33, 392–402.
Mitani, J. C., Gros-Louis, J. & Richards, A. 1996. Sexual dimorphism, the operational sex ratio, and the intensity of male competition in polygynous primates. American Naturalist, 966–980.
Mitra Setia, T. & Van Schaik, C. P. 2007. The response of adult orang-utans to flanged male long calls: inferences about their function. Folia Primatologica, 78, 215–226.
Mitra Setia, T., Delgado, R., Utami Atmoko, S., Singleton, I. & Van Schaik, C. 2009. Social organization and male-female relationships. Orangutans, Geographic Variations in Behavioral Ecology and Conservation, 245–253.
38
Møller, A. P. 1988. Ejaculate quality, testes size and sperm competition in primates. Journal of Human Evolution, 17, 479–488.
Morrogh-Bernard, H. C., Morf, N. V., Chivers, D. J. & Krützen, M. 2011. Dispersal patterns of orang-utans (Pongo spp.) in a Bornean peat-swamp forest. International Journal of Primatology, 32, 362–376.
Muller, M. N., Kahlenberg, S. M., Thompson, M. E. & Wrangham, R. W. 2007. Male coercion and the costs of promiscuous mating for female chimpanzees. Proceedings of the Royal Society B: Biological Sciences, 274, 1009–1014.
Nijman, V. 2005. Hanging in the Balance: An Assessment of trade in Orang-utans and Gibbons in Kalimantan, Indonesia.
Nonacs, P. 2000. Measuring and using skew in the study of social behavior and evolution. The American Naturalist, 156, 577–589.
Nonacs, P. 2003. Measuring the reliability of skew indices: is there one best index? Animal Behaviour, 65, 615–627.
Nsubuga, A. M., Robbins, M. M., Boesch, C. & Vigilant, L. 2008. Patterns of paternity and group fission in wild multimale mountain gorilla groups. American journal of physical anthropology, 135, 263–274.
Ostner, J., Nunn, C. L. & Schülke, O. 2008. Female reproductive synchrony predicts skewed paternity across primates. Behavioral Ecology, 19, 1150–1158.
Palombit, R. A., Seyfarth, R. M. & Cheney, D. L. 1997. The adaptive value of “friendships” to female baboons: experimental and observational evidence. Animal Behaviour, 54, 599–614.
Perry, S. & Manson, J. H. 1995. A comparison of the mating behavior of adolescent and adult female rhesus macaques (Macaca mulatta). Primates, 36, 27–39.
Plavcan, J. M. 2001. Sexual dimorphism in primate evolution. American journal of physical anthropology, 116, 25–53.
Plavcan, J. M. & van Schaik, C. P. 1992. Intrasexual competition and canine dimorphism in anthropoid primates. American Journal of Physical Anthropology, 87, 461–477.
Plavcan, J. M. & Van Schaik, C. P. 1997. Intrasexual competition and body weight dimorphism in anthropoid primates. American Journal of Physical Anthropology, 103, 37–68.
39
Pradhan, G. R. & Van Schaik, C. P. 2009. Why do females find ornaments attractive? The coercion-avoidance hypothesis. Biological journal of the Linnean Society, 96, 372–382.
Pusey, A., Williams, J. & Goodall, J. 1997. The influence of dominance rank on the reproductive success of female chimpanzees. Science, 277, 828–831.
Reichard, U. & Sommer, V. 1997. Group encounters in wild gibbons (Hylobates lar): agonism, affiliation, and the concept of infanticide. Behaviour, 1135–1174.
Schürmann, C. 1981. Courtship and mating behavior of wild orangutans in Sumatra. In: Primate behavior and sociobiology, pp. 130–135. Springer.
Setchell, J. M., Charpentier, M. & Wickings, E. J. 2005. Mate guarding and paternity in mandrills: factors influencing alpha male monopoly. Animal Behaviour, 70, 1105–1120.
Singleton, I. & van Schaik, C. P. 2002. The social organisation of a population of Sumatran orang-utans. Folia Primatologica, 73, 1–20.
Singleton, I., Knott, C. D., Morrogh-Bernard, H. C., Wich, S. A. & van Schaik, C. P. 2009. Ranging behavior of orangutan females and social organization. Orangutans: Geographic variation in behavioral ecology and conservation, 205–212.
Smouse, P. E. & Meagher, T. R. 1994. Genetic analysis of male reproductive contributions in Chamaelirium luteum (L.) gray (Liliaceae). Genetics, 136, 313–322.
Spillmann, B., Dunkel, L. P., Van Noordwijk, M. A., Amda, R. N., Lameira, A. R., Wich, S. A. & Van Schaik, C. P. 2010. Acoustic Properties of Long Calls Given by Flanged Male Orang-Utans (Pongo pygmaeus wurmbii) Reflect Both Individual Identity and Context. Ethology, 116, 385–395.
Sprague, D. S. 1992. Life history and male intertroop mobility among Japanese macaques (Macaca fuscata). International Journal of Primatology, 13, 437–454.
Steenbeek, R., Sterck, E. & De Vries, H. (n.d.).van Hooff JARAM (2000) Costs and benefits of the one-male, age-graded and all-male phases in wild Thomas’s langur groups. Primate males: causes and consequences of variation in group composition. Cambridge University Press, Cambridge, 130–145.
Sterck, E. H. M., Watts, D. P. & van Schaik, C. P. 1997. The evolution of female social relationships in nonhuman primates. Behavioral Ecology and Sociobiology, 41, 291–309.
Stoinski, T., Rosenbaum, S., Ngaboyamahina, T., Vecellio, V., Ndagijimana, F. & Fawcett, K. 2009. Patterns of male reproductive behaviour in multi-male groups of mountain gorillas:
40
examining theories of reproductive skew. Behaviour, 146, 1193–1215.
Strier, K. B., Chaves, P. B., Mendes, S. L., Fagundes, V. & Di Fiore, A. 2011. Low paternity skew and the influence of maternal kin in an egalitarian, patrilocal primate. Proceedings of the National Academy of Sciences, 108, 18915–18919.
Struhsaker, T. T. 1977. Infanticide and social organization in the redtail monkey (Ceraopithecus ascanius schmidti) in the Kibale forest, Uganda. Zeitschrift für Tierpsychologie, 45, 75–84.
Stumpf, R., Thompson, M. E. & Knott, C. 2008. A comparison of female mating strategies in Pan troglodytes and Pongo spp. International Journal of Primatology, 29, 865–884.
Swedell, L. & Tesfaye, T. 2003. Infant mortality after takeovers in wild Ethiopian hamadryas baboons. American journal of primatology, 60, 113–118.
Takahata, Y., Huffman, M. A., Suzuki, S., Koyama, N. & Yamagiwa, J. 1999. Why dominants do not consistently attain high mating and reproductive success: a review of longitudinal Japanese macaque studies. Primates, 40, 143–158.
Thompson, E. 1975. The estimation of pairwise relationships. Annals of human genetics, 39, 173–188.
Tutin, C. E. 1979. Mating patterns and reproductive strategies in a community of wild chimpanzees (Pan troglodytes schweinfurthii). Behavioral Ecology and Sociobiology, 6, 29–38.
Utami, S. & Setia, T. M. 1995. Behavioral changes in wild male and female Sumatran orangutans (Pongo pygmaeus abelii) during and following a resident male take-over. In: The neglected ape, pp. 183–190. Springer.
Utami, S. S., Goossens, B., Bruford, M. W., De Ruiter, J. R. & van Hooff, J. A. 2002. Male bimaturism and reproductive success in Sumatran orang-utans. Behavioral Ecology, 13, 643–652.
Utami Atmoko, S., Singleton, I., Van Noordwijk, M., Van Schaik, C. & Mitra Setia, T. 2009a. Male-male relationships in orangutans. Orangutans: geographic variation in behavioral ecology and conservation. Oxford University Press, New York, 225–234.
Utami Atmoko, S., Setia, T. M., Goossens, B., van Schaik, C. P. & van Noordwijk, M. A. 2009b. Orangutan mating behavior and strategies. Orangutans: Geographic Variation in Behavioral Ecology and Conservation, Oxford University Press, Oxford, 235–244.
41
van Noordwijk, M. A. & van Schaik, C. P. 2005. Development of ecological competence in Sumatran orangutans. American Journal of Physical Anthropology, 127, 79–94.
Van Noordwijk, M. A. & Van Schaik, C. P. 2001. Career moves: transferand rank challenge decisions by male long-tailed macaques. Behaviour, 138, 359–395.
van Schaik, C. P. 1999. The socioecology of fission-fusion sociality in orangutans. Primates, 40, 69–86.
Van Schaik, C. P. 1983. Why are diurnal primates living in groups? Behaviour, 120–144.
Van Schaik, C. P. & Kappeler, P. M. 1997. Infanticide risk and the evolution of male-female association in primates. Proceedings of the Royal Society of London. Series B: Biological
Sciences, 264, 1687–1694.
Van Schaik, C. P., Pandit, S. A. & Vogel, E. R. 2004. A model for within-group coalitionary aggression among males. Behavioral Ecology and Sociobiology, 57, 101–109.
Wang, J. 2012. Computationally efficient sibship and parentage assignment from multilocus marker data. Genetics, 191, 183–194.
Watts, D. P. 1989. Infanticide in mountain gorillas: new cases and a reconsideration of the evidence. Ethology, 81, 1–18.
Watts, D. P. 1991. Mountain gorilla reproduction and sexual behavior. American Journal of Primatology, 24, 211–225.
Watts, D. P. 1998. Coalitionary mate guarding by male chimpanzees at Ngogo, Kibale National Park, Uganda. Behavioral Ecology and Sociobiology, 44, 43–55.
Watts, D. P. 2001. Social relationships of female mountain gorillas. CAMBRIDGE STUDIES IN BIOLOGICAL AND EVOLUTIONARY ANTHROPOLOGY, 215–240.
Wich, S., Utami-Atmoko, S., Setia, T. M., Rijksen, H., Schürmann, C., Van Hooff, J. & Van Schaik, C. 2004. Life history of wild Sumatran orangutans (Pongo abelii). Journal of Human Evolution, 47, 385–398.
Wich, S., Utami-Atmoko, S., Setia, T. M., Djoyosudharmo, S. & Geurts, M. 2006. Dietary and energetic responses of Pongo abelii to fruit availability fluctuations. International Journal of Primatology, 27, 1535–1550.
42
Wickings, E., Bossi, T. & Dixson, A. 2009. Reproductive success in the mandrill, Mandrillus sphinx: correlations of male dominance and mating success with paternity, as determined by DNA fingerprinting. Journal of Zoology, 231, 563–574.
Wrangham, R. 1979. On the evolution of ape social systems. Social Science Information, 18, 336–368.
Wroblewski, E. E., Murray, C. M., Keele, B. F., Schumacher-Stankey, J. C., Hahn, B. H. & Pusey, A. E. 2009. Male dominance rank and reproductive success in chimpanzees, Pan troglodytes schweinfurthii. Animal Behaviour, 77, 873–885.
43
8 APPENDIX
8.1 Appendix: Materials and methods
Table 8-1: Microsatellite loci used in this study
MP Marker ID Repetition pattern Sequence (5’→ 3’) Dye
HM
P1
D1S550 Tetranucleotide F: CCTGTTGCCACCTACAAAAG R: TAAGTTAGTTCAAATTCATCAGTGC
PET
D2S1326 Tetranucleotide F: AGACAGTCAAGAATAACTGCCC R: CTGTGGCTCAAAAGCTGAAT
NED
D3S2459 Tetranucleotide F: CTGGTTTGGGTCTGTTATGG R: AGGGACTTAGAAAGATAGCAGG
VIC
D4S2408 Tetranucleotide F: AATAAACTTCAACTTCAATTCATCC R: AGGTAAAGGCTCTTCTTGGC
FAM
D5S1470 Tetranucleotide F: CATGCACAGTGTGTTTACTGG R: TAGGATTTTACTATATTCCCCAGG
FAM
HM
P2
D13S321 Tetranucleotide F: TACCAACATGTTCATTGTAGATAGA R: CATACACCTGTGGACCCATC
PET
D13S765 Tetranucleotide F: TGTAACTTACTTCAAATGGCTCA R: TTGAAACTTACAGACAGCTTGC
NED
D16S420 Dinucleotide F: ATTTCCTGAGGTCTAAAGCACCC R: TTAGGCCCAGTCCACACTCAAG
VIC
D2S141 Dinucleotide F: ACTAATTACTACCCNCACTCCC R: TTTTCCAAACAGATACAGTGAACTT
FAM
D5S1505 Tetranucleotide F: TAAGTGCCAGAGTCTCCCAC R: TAAGGCATGTCTCGGAGCTA
FAM
D6S501 Tetranucleotide F: GCTGGAAACTGATAAGGGCT R: GCCACCCTGGCTAAGTTACT
PET
PM
P1
D4S1627 Tetranucleotide F: AGCATTAGCATTTGTCCTGG R: GACTAACCTGACTCCCCCTC
PET
O4A5 Tetranucleotide F: ATGGGCCCAGAAAACAACTCAGT R: AGATAAAGGAATGGATAGATGGACAGA
FAM
O4A8 Tetranucleotide F: CACAGGGTCCAAACTCAGATTATTG R: GTTTCCTTCCCCCTCATGTAGTTATCAA
NED
O4B24 Tetranucleotide F: TCTGAGGTACCCTGTAACAAAGAAA R: GAAATCCCAGTACCATATAAATGTCAT
FAM
O4B5 Tetranucleotide F: GAGCCCTGATTCGTTTTACTGG R: AGCAAAGGCAGAAAACTGTAATGA
VIC
O4B6 Tetranucleotide F: TGGAGCCTGAATATGTGACTGAAT R: AATGCCAGGATTTCCTTCTTTTT
FAM
O4_6 Tetranucleotide F: GGCAATGTAACATATCCCTCTGTGT R: AGCCATGGACCTTGTGAGAAAAG
PET
PM
P2
D5S1457 Tetranucleotide F: TAGGTTCTGGGCATGTCTGT R: TGCTTGGCACACTTCAGG
FAM
O4A1 Tetranucleotide F: CTCCCCTTCCTTCCTTTATTCAGTT R: CAACACTTGGCAGTCACAAATCAG
FAM
O4B17 Tetranucleotide F: GTACCGACGGTGCACGAACAATGTA R: AGCCTGGCTGAAAAGTGGAACTGAG
PET
O4B3 Tetranucleotide F: TTCCAGAAGGGGCGAGAAGTT R: GTTGGGACCAAACAGTTGTCAATAA
VIC
O4C13 Tetranucleotide F: CTGGGCACACTGTATATGGGGTAG R: GTTTGAGACCACTCATGATGCAAAGACC
FAM
O4C9 Tetranucleotide F: TGCAGGCCAGGGCTTCTTTCAA R: CAGTCTCCCCAGGACCCCTACACAG
FAM
PM
P3
O4A7 Tetranucleotide F: ACTGGCCCATTCAAAGTCTGTCATT R: ACTGGCCCATTCAAAGTCTGT
VIC
O4B20 Tetranucleotide F: CCTGCATTTTGTCACTCCCTCAACC R: CTGCCACACCTCCATGGACACAGAT
NED
P_TETRA_Chr5 Tetranucleotide F: CAGCAGCTCCTGAAATATCTGTCC R: GTTTGGGGTAGAGGAAAGCAGGTTGAT
PET
P_TETRA_Chr7 Tetranucleotide F: CATCTCTTTATGGCTGACTGTTGAT R: GTTTGGTCCAAGACAAATTTGTATGAGT
NED
44
8.2 Appendix: Results
8.2.1 DNA extraction 76 samples were extracted for a total of 95 extractions performed. DNA quantification was
completed over 5 separate qPCR reactions. The correlation of determination R2 for the five
Standard curves always exceeded 0.95 and averaged 0.9882±0.0049, indicating that the
estimates for the DNA concentration in the extracts are reliable.
Estimates revealed that concentrations of target DNA varied between 0 pg/ul and 3.3 ng/ul
with an average of 151.28 pg/ul (table 8-2). When divided in the categories presented in
Morin 2001, the distribution of DNA concentration appeared to be satisfying, with 40% of the
samples reaching and exceeding 25 pg/ul (figure 8-1). 11 extracts appeared to contain no DNA
but it is possible that a high concentration of PCR inhibitors prevented its amplification and
thus its detection.
Figure 8-1: Distribution of DNA concentration in the extracts according to the 4 categories proposed by Morin et al. 2001
Samples with a concentration above 25pg/μl were used for the genotyping, even though
observation suggested that also extracts with lower concentrations can produce satisfying
0
5
10
15
20
25
30
35
40
45
50
Nu
mb
er o
f sa
mp
les
DNA Concentration (pg/μl)
200+
100-200
25-100
0-25
0
45
results and an attempt at genotyping for the 6 HMP2 loci was made with several of these low-
concentration extracts.
Table 8-2: DNA concentration, repetitions required according to Morin et al. (2001) and number of successfully typed loci for each extract.
Inventory Number Mean DNA conc. Repetitions Loci typed
2077 1.494149 DNU 9
2099 1.204544 DNU 0
4288 12.96265 DNU 18
4293 26.95009 7 18
4314 1.756866 DNU 0
4326 67.07179 7 25
4351 5.724094 DNU 0
4353 0 DNU 0
4355 3.487272 DNU 0
4365 0.599858 DNU 0
4376 1.798685 DNU 0
4702 1.180709 DNU 0
4743 7.245081 DNU 0
5468 5.339406 DNU 0
5469 8.739554 DNU 3
5473 4.456352 DNU 0
5473 10.73956 DNU 18
5474 0 DNU 0
5474 0 DNU 0
5475 1.371463 DNU 0
5475 2.524452 DNU 0
5482 9.195361 DNU 14
5502 1.582077 DNU 0
5502 3.452386 DNU 0
5504 2.741074 DNU 0
5504 4.633104 DNU 0
5507 0 DNU 0
5507 3.912551 DNU 0
5509 8.745291 DNU 0
5510 0 DNU 0
5517 0 DNU 0
5517 4.221138 DNU 0
5521 0 DNU 0
5521 0 DNU 0
5521 1.884094 DNU 0
5522 0 DNU 0
5522 7.12949 DNU 0
5544 3306.575 2 27
5545 166.0757 4 0
5545 171.1551 4 10
5546 330.3779 2 27
5547 1056.113 2 9
5548 182.3777 4 10
5549 301.6974 2 10
5550 11.77066 DNU 0
5550 110.752 4 27
5554 324.3283 2 27
46
Table 8-2: Continued
5556 1.125215 DNU 0
5556 245.8101 2 24
5561 119.7313 4 25
5564 19.45194 DNU 27
5565 0 DNU 0
5565 2.88426 DNU 0
5565 8.958195 DNU 0
5566 26.12448 7 27
5567 10.30438 DNU 10
5569 4.853313 DNU 0
5569 99.22164 7 26
5575 6.722113 DNU 0
5575 29.92129 7
5576 752.7161 2 0
5579 2.632999 DNU
5581 66.39281 7 10
5583 6.400699 DNU 7
5584 9.563138 DNU 0
5584 15.51868 DNU 25
5585 5.604093 DNU 1
5586 117.1504 4 27
5590 473.5454 2 5
5591 151.0537 4 27
5594 145.8735 4 5
5602 624.6607 2 26
5604 640.0832 2 27
5610 2.426561 DNU
5610 5.470359 DNU
5611 78.10532 7 5
5615 1.227626 DNU
5616 22.52458 DNU 27
5617 12.99845 DNU 5
5619 103.4046 4 5
5622 11.58196 DNU 25
5623 0 DNU
5624 64.01858 7 5
5625 427.9674 2 5
5627 19.22052 DNU 18
5629 117.8612 4 16
5630 213.6555 2 16
5636 55.45231 7 9
5641 39.0487 7 26
5649 586.8771 2 26
5715 2282.97 2 0
5719 108.2326 4 5
5720 222.9017 2 27
5732 144.1035 4 26
5736 91.76767 7 9
Average 151.27
47
8.2.2 Genotyping After DNA quantification, 44 extracts were genotyped for the 6 HMP2 loci. Eleven extracts
were duplicates from previously sampled individuals. 33 extracts were then genotyped for up
to 22 additional loci. After encountering difficulties in the allele scoring process, I have
discarded Locus D2S141 from the analysis.
Identity analysis showed that the 44 extracts exhibited 30 individual genotypes. Of these 30
unique genotypes, 9 matched with genotypes already present in the database that were
produced by former members of the lab. This repetition of previously found genotypes
nonetheless resulted to be particularly useful as it allowed to estimate the level of consistency
in the allele scoring between the different people who worked on the creation of this dataset.
The higher quality of several of the more recent samples I have extracted allowed me to
identify some erroneous allele calls which could have heavily impacted the pedigree analysis.
The final dataset for this population includes 57 unique genotypes and field observation and
photos allowed to assign this genotypes to the corresponding individuals.
8.2.3 Error rates Error rates were estimated comparing repeated genotypes and 4 mismatches were found
among 310 allele comparisons, 2 of which were found at one locus in one of the comparisons.
After examination it was obvious that for that particular locus the error was introduced in the
allele calling process and had nothing to do with the quality of the extracts. The error rate was
then estimated to be approximately 1%. For the Tuanan dataset, the error rate had been
estimated to be 0.112% (Arora et al. 2012).
8.2.4 Evaluation of the genetic markers
8.2.4.1 Suaq
The Hardy-Weinberg test revealed that the loci D6S501, D4S1627, O4A7, PON_TETRA_Chr5
and PON_TETRA_Chr7 were not in Hardy-Weinberg equilibrium but after Bonferroni
correction (new threshold for p-values= 0.001852), only locus D4S1627 appeared to be in
disequilibrium and was discarded. Further testing for HWE was performed in GenAlex and
resulted in the exclusion of two additional loci, O4A7 and PON_TETRA_Chr5.
Testing for Linkage Disequilibrium using Genepop showed that no genetic disequilibrium was
found in between the remaining tested loci. A significance-threshold of 0.001852 for the p-
values was used. Additionally, locus O4B20 was found to be monomorphic and was also
removed from the set of genetic markers used in the downstream analysis. The resulting and
final dataset was therefore reduced to 23 loci.
48
8.2.4.2 Tuanan
After testing for Hardy-Weinberg disequilibrium I have removed locus O4A8 from the dataset.
Loci 04B2O and O4B3 resulted to be uninformative and were also excluded and not used for
the rest of the analysis. The final dataset comprises 25 loci.
8.2.5 Investigation of descriptive genetic parameters and analytical power In the final step before paternity analysis, I have performed a series of tests aimed at
assessing the suitability of both my datasets for paternity analysis. Results for the number of
allele per locus (NA), expected (HE) and observed (HO) heterozygosity, Polymorphic Information
Content (PIC) and probability of identity for both unrelated individuals (Pi) and siblings (Pis) are
summarized in tables 8-2 and 8-3. Figure 8-2 and 8-3 show how Pi and Pis decrease by
including more loci in the analysis
Table 8-2 Genetic parameters for the markers used for the Suaq population. N=Number of individuals
genotyped, NA= Number of alleles for the marker, Ho= Observed Heterozygosity, HE= Expected
Heterozygosity, PIC= Polymorphic Information Content, Pi= Probability of identity, Pis = Probability of
identity (siblings), in italic: HMP2 loci
Locus N NA HO HE NE1P NE2P PIC Pi Pis
D1S550 56 6 0.839 0.775 0.632 0.454 0.729 0.092 0.389
D2S1326 56 3 0.339 0.374 0.931 0.808 0.339 0.428 0.672
D3S2459 57 4 0.772 0.708 0.722 0.553 0.649 0.142 0.434
D4S2408 53 5 0.698 0.667 0.768 0.613 0.596 0.18 0.465
D5S1470 55 5 0.745 0.692 0.733 0.563 0.635 0.149 0.445
D13S321 55 8 0.836 0.75 0.657 0.481 0.702 0.107 0.405
D13S765 55 6 0.6 0.652 0.768 0.615 0.582 0.19 0.474
D16S420 54 8 0.796 0.823 0.543 0.368 0.789 0.06 0.357
D5S1505 56 7 0.607 0.723 0.691 0.517 0.671 0.126 0.423
D6S501 56 4 0.5 0.632 0.798 0.646 0.561 0.205 0.488
O4A5 53 4 0.604 0.549 0.85 0.733 0.46 0.292 0.551
O4A8 55 3 0.218 0.242 0.971 0.886 0.218 0.599 0.78
O4B24 55 3 0.509 0.531 0.862 0.721 0.469 0.282 0.557
O4B5 54 6 0.796 0.745 0.672 0.497 0.694 0.113 0.409
O4B6 50 6 0.52 0.501 0.862 0.698 0.469 0.281 0.572
O4_6 55 4 0.582 0.675 0.755 0.596 0.608 0.171 0.458
D5S1457 56 6 0.768 0.711 0.701 0.521 0.666 0.126 0.429
O4A1 54 5 0.833 0.695 0.739 0.575 0.631 0.155 0.444
O4B17 54 6 0.741 0.688 0.743 0.579 0.624 0.159 0.449
O4B3 54 4 0.63 0.632 0.801 0.656 0.555 0.211 0.49
O4C13 49 4 0.755 0.626 0.805 0.668 0.541 0.223 0.496
O4C9 51 4 0.588 0.501 0.868 0.707 0.464 0.286 0.573
PON_7 42 13 0.786 0.835 0.494 0.324 0.81 0.046 0.349
Mean 53.6956 5
5.391304
0.65487
0.640304
1.1E-3
4.3E-06
0.585304 kk
3.2E-16
1.8E-07
49
Table 8-3 Genetic parameters for the markers used for the Tuanan population. N=Number of individuals genotyped, NA= Number of alleles for the marker, Ho= Observed Heterozygosity, HE= Expected Heterozygosity, PIC= Polymorphic Information Content, Pi= Probability of identity, Pis = Probability of identity (siblings), in italic: HMP2 loci
Locus N NA HO HE NE1P NE2P PIC Pi Pis
D1S550 40 7 0.65 0.709 0.71 0.541 0.65 0.14 0.435
D2S1326 45 3 0.556 0.521 0.867 0.731 0.456 0.294 0.566
D3S2459 43 6 0.837 0.812 0.57 0.392 0.774 0.068 0.366
D4S2408 39 5 0.744 0.653 0.759 0.583 0.604 0.166 0.469
D5S1470 44 4 0.477 0.499 0.874 0.736 0.441 0.309 0.581
D13S321 56 5 0.518 0.648 0.785 0.636 0.573 0.197 0.478
D13S765 55 5 0.6 0.658 0.77 0.617 0.586 0.187 0.471
D16S420 53 5 0.528 0.508 0.87 0.731 0.45 0.3 0.573
D2S141 54 8 0.833 0.769 0.634 0.457 0.724 0.095 0.393
D5S1505 53 8 0.755 0.77 0.618 0.438 0.732 0.087 0.39
D6S501 51 8 0.745 0.722 0.687 0.509 0.675 0.121 0.423
D4S1627 36 6 0.694 0.644 0.763 0.586 0.598 0.17 0.475
O4A5 38 6 0.658 0.676 0.737 0.559 0.628 0.15 0.454
O4B24 20 2 0.05 0.05 0.999 0.976 0.048 0.906 0.952
O4B5 41 5 0.659 0.619 0.791 0.626 0.564 0.199 0.494
O4B6 37 7 0.405 0.747 0.652 0.47 0.705 0.101 0.407
O4_6 42 3 0.738 0.618 0.813 0.674 0.535 0.227 0.501
D5S1457 42 7 0.857 0.785 0.618 0.44 0.74 0.086 0.384
O4A1 40 4 0.825 0.751 0.682 0.507 0.693 0.115 0.408
O4B17 37 7 0.73 0.802 0.581 0.402 0.763 0.072 0.373
O4C13 32 4 0.406 0.567 0.842 0.736 0.46 0.293 0.544
O4C9 39 4 0.564 0.647 0.78 0.627 0.574 0.195 0.479
O4A7 39 2 0.333 0.315 0.952 0.869 0.262 0.523 0.726
PON_TETRA_Chr5
36 6 0.806 0.786 0.619 0.44 0.739 0.087 0.384
PON_TETRA_Chr7
34 15 0.912 0.91 0.345 0.208 0.888 0.019 0.307
Mean 41.8
5.68 0.635 0.64844
0.647
0.64844
2.5E-4
4E-07
0.594 6E-21
5.5E-09
50
Figure 8-2: Cumulative probability of identity for Tuanan
Figure 8-3: Cumulative probability of identity for Suaq
1E-21
1E-19
1E-17
1E-15
1E-13
1E-11
1E-09
1E-07
1E-05
0.001
0.1
0 5 1 0 1 5 2 0 2 5 3 0
PI (
LOG
AR
YTH
MIC
SC
ALE
)
CUMULATIVE NUMBER OF LOCI
Pi_TUA
Pis_TUA
1.E-18
1.E-16
1.E-14
1.E-12
1.E-10
1.E-08
1.E-06
1.E-04
1.E-02
1.E+00
0 5 1 0 1 5 2 0 2 5
PI (
LOG
AR
YTH
MIC
SC
ALE
)
CUMULATIVE NUMBER OF LOCI
Pi_SUA
Pis_SUA
51
Figures 8-4 and 8-5 shows the results for the analysis of informativeness with KinInfor. These
results suggest that for both populations, my dataset is powerful enough for the identification
of parent offspring pairs, even in presence of full siblings. (74% of successfully discriminated
pairs at 95% significance level for Suaq, 82% for Tuanan).
The results of this analysis also show that without a basic initial pedigree reconstruction, it
would be difficult to identify half-sib pairs with necessary confidence. In fact, in approximately
30% of the cases, my dataset won’t be strong enough to discriminate between half-sibs and
unrelated individuals. This problem accentuates when other distantly related individuals are
present in the population. For this reason I have decided that it will be necessary to initially
identify parent offspring, the most easily identified relationship with my dataset, and then
proceed in building an initial pedigree with these findings. The identification of half-sibship for
pairs in which at least one of the four parents is known will be easier.
Figure 8-4: Estimates of PWR values for 4 hypotheses pairs for Suaq. HS=Half-siblings, FC=First
cousins, U=Unrelated, PO=Parent-offspring, FS=Full siblings).
0.00
0.10
0.20
0.30
0.40
0.50
0.60
0.70
0.80
0.90
1.00
HS-FC HS-U PO-FS PO-U
PW
r
Relationship (Δ1-Δ2)
PWR_80_SUA
PWR_95_SUA
52
Figure 8-5: Estimates of PWR values for 4 hypotheses pairs for Tuanan. HS=Half-siblings, FC=First
cousins, U=Unrelated, PO=Parent-offspring, FS=Full siblings)
8.2.6 Parentage analysis: Parameters used
Table 8-4: Parameters used for the parentage analysis in Cervus
Analysis parameters in Cervus 3.0 Maternity Paternity
Suaq Tuanan Suaq Tuanan
Number of offspring 10000 - 10000 10000
Number of candidate mothers/fathers 45 - 100 100
Proportion of mothers/fathers sampled 0.5 - 0.25 0.25
Proportion of loci typed 0.94 - 0.94 0.755
Proportion of loci mistyped 0.01 - 0.01 0.01
Error rate 0.01 - 0.01 0.01
Confidence determined using Delta - Delta Delta
Relaxed confidence level 80% - 80% 80%
Strict confidence level 95% - 95% 95%
0.00
0.10
0.20
0.30
0.40
0.50
0.60
0.70
0.80
0.90
1.00
HS-FC HS-U PO-FS PO-U
PW
r
Relationship (Δ1-Δ2)
PWR_80_TUA
PWR_95_TUA
53
Table 8-5: Parameters used for the parentage analysis in COLONY
Analysis parameters in COLONY 2.0 Maternity Paternity
Suaq Tuanan Suaq Tuanan
Number of loci 23 25 23 26
Number of offspring in the sample 18 18 22 27
Outbreeding (0) or inbreeding (1) model 0 0 0 0
Number of male candidates 0 0 23 23
Number of female candidates 18 10 18 9
Number of known paternal sibships 0 0 0 0
Number of known maternal sibships 0 0 0 9
Number of offspring with excluded fathers 0 0 0 0
Number of offspring with excluded mothers 0 0 0 4
Male mating system Polygamy Polygamy Polygamy Polygamy
Female mating system Polygamy Polygamy Polygamy Polygamy
Number of Excluded Paternal Sibships 0 0 0 0
Number of Excluded Maternal Sibships 0 0 0 0
Seed for random number generator 1234 1234 1234 1234
Allele frequency Update Update Update Update
Sibship complexity prior Yes Yes Yes Yes
Length of run Long Long Very Long Long
Likelihood High Very High Very High Very High
Probability a mum is included in the female candidates 0.6 0.6 0.6 0.6
Probability a dad is included in the male candidates - - 0.25 0.25
Error rate 0.01 0.01 0.01 0.01
54
8.2.7 Maternity: Suaq Maternity analysis in Cervus 3.0 and COLONY 2.0 revealed 17 mother-offspring pairs (Table 8-
6).
Table 8-6: Results of maternity analysis performed with Cervus 3.0 and COLONY 2.0 (1Pne= First parent exclusion probability, 2Pne=Second parent exclusion probability, LC= Number of loci compared, LMM= Number of mismatches, Pair confidence= *=95%, +=80%)
Cervus COLONY
Offspring ID
1Pne Mother
ID LC LMM
Pair LOD
Pair
ΔLOD
Pair confidence
Likelihood Pair
confidence
CHICK 9.46E-04 DODI 21 0 8.14 8.14 * 1.000 *
CHINDY 4.70E-03 CISSY 22 0 7.82 6.46 * 1.000 *
DIDDY 4.32E-04 DODI 19 0 8.62 7.24 * 1.000 *
DODI 2.26E-03 CHICK 21 0 8.14 8.14 * 1.000 *
ELLIE 8.32E-03 FRISKA 22 0 8.01 8.01 * 1.000 *
LENA 5.97E-04 DIANA 23 11 5.25 5.25 * 1.000 *
LILLY 1.51E-03 LISA 23 0 4.74 4.70 * 1.000 *
LISA 5.98E-03 CISSY 22 0 6.35 2.65 * 1.000 *
NIBLA 5.04E-04 INTAY 23 0 9.52 9.52 * 1.000 *
RAFFI 8.61E-04 HALTE 20 0 7.16 5.80 * 1.000 *
ROBBI 2.27E-04 LENA 23 0 13.0 10.6 * 1.000 *
SHERA 1.55E-03 CHICK 23 0 5.26 5.26 * 1.000 *
HERDY 1.42E-03 FRISKA 22 0 6.62 6.62 * 1.000 *
TEDI 2.12E-04 HALTE 22 12 4.74 3.23 * 1.000 *
TINA 5.11E-03 RAFFI 20 0 5.81 5.81 * 1.000 *
UNO 2.13E-03 BIB 21 0 9.56 9.56 * 1.000 *
HALTE 3.32E-03 FRISKA 22 0 4.58 4.58 * 1.000 *
FREDDY 2.92E-04 FRISKA 22 0 6.84 1.88 + 3 1.000 * 1, 2 Mismatch likely caused by genotyping error 3Requirements for significance at strict confidence level were not met due to the presence of a female relative, probably on the paternal side, in the pool of candidate mothers (see Marshall et al. 1998, Kalinowsky 2007). Field observations allowed to confirm this relationship
8.2.8 Paternity concentration: Suaq Table 8-8: Results of the reproductive skew analysis for Suaq with Skew Calculator 2013 for 8
estimators (Nonacs 2003).
Number of males
Number of successful
males S Sc S3 Q λ Iδ Ip B P level
44 8 0.8317 0.2386 0.9009 0.0884 0.3194 4.8 0.552 0.0464 0.048
55
8.2.9 Maternity: Tuanan Maternity analysis for post-dispersal individuals with Cervus 3.0 was already completed by
former lab members (Arora et al. 2012). I performed a maternity analysis including juvenile
individuals in COLONY and identified 15 mother-offspring pair (Table 8-7).
Table 8-7: results of the maternity analysis performed in COLONY 2.0 for Tuanan.
Offspring ID Mother ID Confidence level Significance
(*) = 0.95 (+) = 0.80
CIKIPOS WILMA 1.000 *
DESY INUL 1.000 *
FRODO WILMA 1.000 *
IDO INUL 1.000 *
MILO MINDY 1.000 *
PUMUCKL PINKY 1.000 *
STREISEL SIDONY 0.999 *
SUSI SUMI 0.977 *
JIP JUNI 0.998 *
MINDY JINAK 0.997 *
JERY JINAK 0.988 *
KONDOR KERRY 0.989 *
KERRY JINAK 0.999 *
LOLO SIDONY 0.985 *
JUNI JINAK 0.988 *
TIKUS TALIA 0.982 *
top related